Moore v. United States En Banc

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 No. 19-CF-0687 BRIAN E. MOORE, APPELLANT, v. UNITED STATES, APPELLEE. Appeal from the Superior Court of the District of Columbia (2018-CF3-011411) (Hon. Craig Iscoe, Motions Judge) (Hon. Milton C. Lee, Jr., Trial Judge) (Argued En Banc February 29, 2024 Decided September 4, 2025) Sean R. Day for appellant. William Collins, Public Defender Service, with whom Samia Fam and Jaclyn S. Frankfurt, Public Defender Service, were on the brief, as amicus curiae in support of appellant. Chrisellen R. Kolb, Assistant United States Attorney, with whom Matthew M. Graves, United States Attorney at the time the brief was filed, and John P. Mannarino and Katherine M. Kelly, Assistant United States Attorneys, were on the brief, for appellee. Graham E. Phillips, Deputy Solicitor General, with whom Brian L. Schwalb, Attorney General for the District of Columbia, Caroline S. Van Zile, Solicitor General for the District of Columbia, and Ashwin P. Phatak, Principal Deputy Solicitor General, were on the brief, as amicus curiae in support of appellee. 2 Before BLACKBURNE-RIGSBY, Chief Judge, and BECKWITH, EASTERLY, MCLEESE, DEAHL, HOWARD, and SHANKER, Associate Judges. Opinion for the court by Associate Judge SHANKER, with whom BLACKBURNE-RIGSBY, Chief Judge, and MCLEESE and DEAHL, Associate Judges, join. Concurring opinion by Associate Judge MCLEESE at page 42. Concurring opinion by Associate Judge DEAHL at page 47. Dissenting opinion by Associate Judge EASTERLY, with whom BECKWITH and HOWARD, Associate Judges, join, at page 61. SHANKER, Associate Judge: Appellant Brian E. Moore twice told his attorney, John Harvey, that he was going to kill the District of Columbia Assistant Attorney General (AAG) prosecuting a criminal contempt case against him. First, Mr. Moore stated that he was going to “shoot that bitch,” adding that he in fact owned guns. Then, about two months later, and after Mr. Harvey had warned Mr. Moore that he would take future threats seriously, Mr. Moore repeatedly exclaimed that he would “bust a cap in this bitch,” again referring to the prosecutor. In response, Mr. Harvey sought to withdraw from representing Mr. Moore and, upon a court order, disclosed Mr. Moore’s threats. Mr. Harvey testified before a grand jury, which indicted Mr. Moore; and he testified again at Mr. Moore’s trial, at the conclusion of which a jury convicted Mr. Moore for threatening the AAG and obstructing justice. On appeal, Mr. Moore argues that Mr. Harvey’s testimony was inadmissible because his statements were protected by the attorney-client privilege. We disagree 3 and affirm the Superior Court’s admission of Mr. Harvey’s testimony. We hold that criminal threats to cause death or substantial bodily harm fall outside the attorney-client privilege. I. Background A. Factual Background The evidence at trial supported the following. Mr. Harvey represented Mr. Moore in a criminal contempt proceeding arising out of the allegation that Mr. Moore violated a civil protective order by contacting his then-wife. On April 12, 2018, during the criminal contempt trial, the AAG asked the court to place Mr. Moore on GPS monitoring with an ankle bracelet. Mr. Harvey had previously persuaded the trial court to remove the GPS monitoring, but the AAG sought reconsideration of that ruling. The AAG’s renewed attempts to impose GPS monitoring frustrated Mr. Moore. According to Mr. Harvey, his conversation with Mr. Moore went as follows. When they stepped into the hallway during a break in the proceedings, Mr. Moore told Mr. Harvey: “Fuck that bitch. I hate this bitch.” In Mr. Harvey’s experience—consisting of over thirty years as a criminal defense attorney—this kind of anger was normal, and he explained to Mr. Moore that the AAG was just doing 4 her job. Unassuaged, Mr. Moore continued, “Man, fuck that bitch. Fuck that bitch. I’ll shoot that bitch. Fuck that bitch.” “Man, what are you talking about?” Mr. Harvey asked. Mr. Moore clarified, “That’s right, Harvey. I’ll shoot that bitch.” Growing concerned, Mr. Harvey warned, “Man, I’m taking—you starting to make me think you serious.” Mr. Moore reiterated, “God damn right, Harvey. Fuck that bitch. I’ll shoot that bitch.” In case the message was not clear, Mr. Moore added that he in fact owned guns—at least according to Mr. Harvey’s grand-jury testimony, which Mr. Harvey did not repeat at trial. At this point, Mr. Harvey tried to withdraw from the representation. He told Mr. Moore that he could no longer “be a part of this” and was “going to have to withdraw.” “I don’t give a fuck what you do, Harvey,” answered Mr. Moore. “I don’t give a fuck.” Mr. Harvey reached out to “Bar Counsel” and asked what he should do. According to Mr. Harvey, he learned that he was permitted to disclose the threats to the court but that the decision was up to him. Mr. Harvey decided not to disclose Mr. Moore’s statements. He nevertheless asked the court to let him withdraw from the representation. He invoked D.C. R. Pro. Conduct 1.16(b), which permits attorneys to “withdraw from representing a client if,” among other things, “[t]he client persists in a course of action involving 5 the lawyer’s services that the lawyer reasonably believes is criminal or fraudulent” or “[t]he client has used the lawyer’s services to perpetrate a crime or fraud.” D.C. R. Pro. Conduct 1.16(b). Mr. Harvey did not invoke Rule 1.6, which permits, but does not require, attorneys to disclose confidential communications when they believe disclosure is reasonably necessary “to prevent a criminal act . . . likely to result in death or substantial bodily harm absent disclosure.” Id. R. 1.6(c). Mr. Harvey testified that, “at that point, [he] had not reached a decision to reveal the information.” The trial court declined Mr. Harvey’s request to withdraw, because Mr. Harvey refused to explain why he wanted to withdraw and what Mr. Moore had said. Uncertain how to proceed, Mr. Harvey spoke with Mr. Moore again. Mr. Moore recanted his earlier statements and explained that he “was just bullshitting” and “didn’t mean it.” Mr. Harvey warned that he would believe Mr. Moore if he threatened to shoot someone again. “I won’t say nothing like that again,” Mr. Moore promised. “I was just bullshitting.” Mr. Harvey continued representing Mr. Moore. The trial paused for about two months due to scheduling issues, and when it resumed on June 29 the AAG raised new concerns about Mr. Moore and once again requested ankle monitoring. The trial court agreed to place Mr. Moore under GPS 6 tracking to ensure his compliance with court orders. Because the proceeding wrapped up late, there was no time to fit Mr. Moore with the ankle bracelet that day. Worse still, because it was a Friday, Mr. Moore would have to be in the District on Monday morning to fit the tracker. The situation angered Mr. Moore. He had a training seminar for a new job Monday morning in North Carolina, and he feared he would miss the training and lose the job as a result. Upon entering the hallway outside the courtroom, Mr. Moore—according to Mr. Harvey’s testimony—said: “Harvey, if I lose my job, I’m going to bust a cap in this bitch[;] I’m going to bust a cap in this bitch.” “Man, what are you doing?” asked Mr. Harvey. “Man, fuck this bitch. If I lose my job, I’m going to bust a cap in this bitch,” Mr. Moore repeated. He then made a shooting gesture, which a security camera caught. “I told you what I was going to do if you ever said something like that to me again,” warned Mr. Harvey. “Fuck her. Fuck you,” retorted Mr. Moore. Mr. Harvey approached the court and again asked to withdraw. He told the court that he could no longer represent Mr. Moore and that, if the court ordered him to explain why, he would. The court ordered him to disclose the reason. Mr. Harvey told the court that Mr. Moore had threatened to shoot the AAG. The court directed a marshal to take Mr. Moore into custody. A few days later, despite expressing some 7 skepticism about the sincerity of Mr. Moore’s threats, the court granted Mr. Harvey’s request to withdraw. It granted the request largely because Mr. Harvey had represented that his relationship with Mr. Moore had deteriorated to the point where he could no longer adequately represent Mr. Moore. B. Procedural Background The United States impaneled a grand jury to inquire into Mr. Moore’s alleged conduct. The government called Mr. Harvey as a witness. As far as the record reflects, Mr. Harvey did not invoke the attorney-client privilege on Mr. Moore’s behalf, and Mr. Moore had no opportunity to invoke the privilege himself. Mr. Harvey testified about his two hallway conversations with Mr. Moore and the threatening statements. The grand jury charged Mr. Moore with two counts of obstruction of justice, D.C. Code § 22-722(a)(5), and two counts of threatening a public official, D.C. Code § 22-851(c). The two counts for each offense corresponded to the two days on which Mr. Moore made the threats: April 12 and June 29. The case was assigned to a different judge from the one who presided over Mr. Moore’s criminal contempt case. At a pretrial hearing, Mr. Moore moved to exclude Mr. Harvey’s testimony because the attorney-client privilege rendered the testimony inadmissible. He also denied making any threatening statements and testified that “Mr. Harvey flat[ ]out 8 lied.” Assuming the truth of Mr. Harvey’s testimony, the trial court concluded that neither Mr. Moore’s April 12 nor June 29 threats were made for the purpose of obtaining legal assistance. The statements were, according to the court, unrelated “to anything except the desire to kill the prosecutor[,] and that is not a legal purpose in the sense of seeking legal advice.” The court emphasized that “the repeated nature of the statements made clear that [they were] not just ill-advised[;] [they were] well thought out.” Moreover, Mr. Harvey’s warning to Mr. Moore that he would disclose the statements if Mr. Moore repeated them was also powerful evidence to the trial court that the statements were not related to a legal purpose. Accordingly, the trial court concluded that Mr. Harvey’s testimony was not privileged. The case proceeded to trial, during which Mr. Harvey testified as described above. After hearing Mr. Harvey’s testimony and seeing the security-camera footage of the June 29 hallway conversation, the jury convicted Mr. Moore on all counts. The trial court sentenced Mr. Moore to eight years in prison to be followed by five years of supervised release. It also assessed a $400 fee under the Victims of Violent Crime Compensation Act. Mr. Moore appealed his convictions. He raised four types of arguments on appeal: (1) the attorney-client privilege protected Mr. Moore’s statements to Mr. Harvey and thus those statements were inadmissible; 9 (2) the government presented insufficient evidence to convict him because it did not prove that he intended for his threats to reach the target; (3) the trial court improperly instructed the jury; and (4) the trial court infringed on Mr. Moore’s right to testify in his own defense. A divided division of this court vacated Mr. Moore’s convictions, addressing only the first two arguments: the sufficiency of the evidence and the attorney-client privilege. Moore v. United States, 285 A.3d 228, 232 (D.C. 2022). The full division concluded that the government presented sufficient evidence to support the convictions, but the majority vacated Mr. Moore’s convictions on the ground that his statements to Mr. Harvey were privileged and the error in admitting them was not harmless. Id. Judge Thompson dissented because, in her view, Mr. Moore did not utter his threats for the purpose of obtaining legal services and thus the privilege did not apply to them. Id. at 253 (Thompson, J., dissenting). We granted the United States’ petition for en banc review and vacated the division’s decision. Moore v. United States, No. 19-CF-0687, 2023 WL 3674377, at *1 (D.C. May 25, 2023) (per curiam). Because the government’s petition, amicus curiae’s brief in support of the petition, and Mr. Moore’s opposition to the petition all addressed only the attorney-client privilege, we ordered the parties to brief and argue only that issue. Moore v. United States, No. 19-CF-0687 (D.C. June 13, 2023) (order). We therefore reinstate Part I of the division opinion, which concerned the 10 sufficiency of the evidence against Mr. Moore. We address today only the attorney-client-privilege issue and return the case to the division for consideration of Mr. Moore’s remaining arguments regarding the jury instructions and his right to testify. II. Analysis Mr. Moore asserts that the threats against the prosecutor that he expressed to Mr. Harvey were protected by the attorney-client privilege and, therefore, Mr. Harvey’s testimony about the threats should not have been admitted into evidence. Reviewing de novo because we are addressing the scope of the attorney-client privilege, see In re Pub. Def. Serv., 831 A.2d 890, 898 (D.C. 2003), we disagree. We hold that the attorney-client privilege excludes from its protection criminal threats to cause death or serious bodily harm. The attorney-client privilege “is not absolute.” Adams v. Franklin, 924 A.2d 993, 999 (D.C. 2007). We have declined to apply the privilege where it would exceed, or defeat, its purpose. See id. at 998. Because the privilege is a creature of common law in the District, In re Ti.B., 762 A.2d 20, 27-28 (D.C. 2000), we need not mechanically apply it in every circumstance where the traditional elements of the privilege are met. Rather, it falls to us to construe the privilege, and we may recognize exceptions where appropriate. Cf. Perez Hernandez v. United States, 286 11 A.3d 990, 996 (D.C. 2022) (en banc) (rejecting the view that the District’s common law is frozen). Indeed, courts have acknowledged numerous exceptions to the attorney-client privilege over the years. See generally Edward J. Imwinkelried, The New Wigmore: Evidentiary Privileges § 6.13.2 (2d ed. 2009) (discussing exceptions to evidentiary privileges, including the crime/fraud exception and self-defense exception). Thus, we decline to rely on a conclusion that serious, objectively credible 1 threats do not meet the elements of the attorney-client privilege—for example, because they necessarily are not related to the legal representation or do not seek legal advice, or because a client who utters a threat to cause death or serious bodily harm can have no reasonable expectation of confidentiality in their communications. Nor do we find it necessary to decide whether all serious threats uttered to an attorney fall under the crime/fraud exception to the privilege. Instead, we hold that a client’s commission of a new, completed, serious threat crime qualifies as an abuse of the privilege that justifies an exception to the privilege’s ordinary application. 1 We use this “objectively credible” label to refer to a key element of the actus reus of a threat crime: that “the ordinary hearer [would] reasonably . . . believe that the threatened harm would take place.” Carrell v. United States, 165 A.3d 314, 319-20 (D.C. 2017) (en banc) (alterations in original) (internal quotation marks omitted). 12 A. The Attorney-Client Privilege Generally The attorney-client privilege constitutes a “narrow exception[ ] to the general rule that every person must offer testimony upon all facts relevant to a judicial proceeding.” Edmund J. Flynn Co. v. LaVay, 431 A.2d 543, 551 (D.C. 1981); see also Trump v. Vance, 140 S. Ct. 2412, 2420 (2020) (“In our judicial system, ‘the public has a right to every man’s evidence.’” (quote attributed to Lord Chancellor Hardwicke)). Typically, we assume that increased access to relevant evidence enhances the truth-seeking function of judicial proceedings, yielding more accurate and just results. See United States v. Nixon, 418 U.S. 683, 709 (1974) (“The very integrity of the judicial system and public confidence in the system depend on full disclosure of all the facts, within the framework of the rules of evidence. To ensure that justice is done, it is imperative to the function of courts that compulsory process be available for the production of evidence needed either by the prosecution or by the defense.”). That is why evidentiary privileges, which prevent witnesses from offering relevant—sometimes critical—evidence, in some sense operate “in derogation of the search for truth.” Id. at 710. Nevertheless, American jurisdictions tolerate this “derogation” because of the countervailing policies underlying privileges. Commentators and courts justify the attorney-client privilege by, for instance, arguing (1) that fostering candor between 13 attorney and client leads to the more efficient, fair, and accurate administration of justice, see United States v. Zolin, 491 U.S. 554, 562 (1989); (2) that certain relationships, including the relationship between attorney and client, fall within “privacy enclaves” into which the state and opposing parties cannot penetrate, see 1 Kenneth S. Broun et al., McCormick on Evidence § 72 (9th ed.); or (3) that a lawyer testifying against his client constitutes a moral betrayal, 24 Charles A. Wright et al., Federal Practice and Procedure § 5472 (1st ed.). But the endorsement by courts—including this court—of the above policies is not absolute; the attorney-client privilege applies only where the communication at issue meets specific requirements. We have generally followed the Wigmore test, which provides that: (1) where legal advice of any kind is sought (2) from a professional legal advisor in his capacity as such, (3) the communications relating to that purpose, (4) made in confidence (5) by the client, (6) are at his instance permanently protected (7) from disclosure by himself or by the legal adviser, (8) except the protection be waived. Jones v. United States, 828 A.2d 169, 175 (D.C. 2003) (quoting 8 John Henry Wigmore, Evidence in Trials at Common Law § 2292 (McNaughton Rev. 1961)). And even where these requirements are met, the privilege is still subject to a variety of exceptions. For instance, although this court has not yet addressed such a situation, many courts recognize an exception to the attorney-client privilege where 14 an attorney brings an action against their client for lack of payment. E.g., Squires Sanders & Dempsey, L.L.P. v. Givaudan Flavors Corp., 937 N.E.2d 533, 541 (Ohio 2010); see 8 Wigmore, supra, § 2312; Imwinkelried, supra, § 6.13.2(a). In addition, and closest to this case, we exempt from the attorney-client privilege at least some “communications made for the purpose of getting advice for the commission of a fraud or crime.” In re Pub. Def. Serv., 831 A.2d at 901-02 (quoting Zolin, 491 U.S. at 563). So the attorney-client privilege does not encompass all attorney-client communications, and even those communications which it does cover may be subject to exceptions pursuant to overriding policy concerns. Those overriding policy concerns exist, in our view, in the context of criminal threats. Before explaining why, some background on these threats is in order. B. Criminal Threats To commit a threat crime, 2 a defendant must engage in a “defined actus reus”; he must “(1) utter[ ] words to another person (2) with a result that the ordinary hearer 2 Many of this court’s threats cases involve a different threats statute than that used to convict Mr. Moore; they concern either the misdemeanor threats statute, D.C. Code § 22-407, or the felony threats statute, D.C. Code § 22-1810. See Gray v. United States, 100 A.3d 129, 132 n.1 (D.C. 2015) (as amended); Carrell, 165 A.3d at 319-20 (discussing the two threats statutes). Mr. Moore, however, was convicted 15 [would] reasonably . . . believe that the threatened harm would take place.” Carrell, 165 A.3d at 319-20 (first alteration added) (internal quotation marks and footnote omitted). And the act alone is not sufficient; the defendant must act either (1) “with the purpose to threaten,” (2) “with knowledge that his words would be perceived as a threat,” or perhaps (3) with reckless disregard for how his words would be perceived. Id. at 324. Despite constituting a communication, a criminal threat is itself a completed crime. See Evans v. United States, 779 A.2d 891, 894 (D.C. 2001) (contrasting criminal threats with attempted criminal threats). Indeed, unlike an attempt crime, one need not intend to follow through on the threat for it to be criminal. See Carrell, 165 A.3d at 323; cf. United States v. Resendiz-Ponce, 549 U.S. 102, 106 (2007) (observing that attempt crimes require proof that the defendant “intended to commit the completed offense”). We proscribe threats separately from attempts because threats impart additional harms beyond the injury of mere attempt crimes. See Rogers v. United States, 422 U.S. 35, 46-47 (1975) (Marshall, J., concurring) under D.C. Code § 22-851(c), which specifically prohibits threats against public officials. Id. (“A person who stalks, threatens, assaults, kidnaps, or injures any official or employee . . . while the official or employee is engaged in the performance of his or her duties or on account of the performance of those duties . . . .”). Whatever differences may exist between these statutes, we assume for the purposes of our analysis that what it means to issue a threat is the same under each statute. 16 (“Plainly, threats may be costly and dangerous to society in a variety of ways, even when their authors have no intention whatever of carrying them out.”). These harms include sowing fear and intimidating others into acquiescing to demands. See Virginia v. Black, 538 U.S. 343, 359-60 (2003) (“[A] prohibition on true threats ‘protect[s] individuals from the fear of violence’ and ‘from the disruption that fear engenders,’ in addition to protecting people ‘from the possibility that the threatened violence will occur.’” (second alteration in original) (quoting R.A.V. v. City of St. Paul, 505 U.S. 377, 388 (1992))); Jennifer E. Rothman, Freedom of Speech and True Threats, 25 Harv. J.L. & Pub. Pol’y 283, 290-93 (2001) (enumerating the rationales for proscribing threats). Such harms are not limited to the target of the threat, i.e., the person whom the defendant states they will harm, but instead extend to all who hear the threat. One who hears someone utter an objectively credible threat against the life and limb of another is likely to be intimidated themselves. See, e.g., United States v. Ivers, 967 F.3d 709, 716 (8th Cir. 2020) (noting that two attorneys who heard a defendant’s threat “were frightened by what [the defendant] had said” even though the threat was not directed at them); State v. Taupier, 193 A.3d 1, 33 (Conn. 2018) (recognizing that a third party who heard a threat targeted at someone else feared for her own safety). Indeed, in at least partial recognition of this principle, a threat need not necessarily reach its target to be punishable. See Moore, 285 A.3d at 236 (rejecting 17 Mr. Moore’s argument that he could not be convicted of uttering a criminal threat unless the government proved that he specifically intended for his threat to reach its target); United States v. Baish, 460 A.2d 38, 42 (D.C. 1983) (explaining that the threatening message must have been “conveyed to someone—either to the object of the threat or to a third party” (emphasis added)); United States v. Khan, 937 F.3d 1042, 1051 (7th Cir. 2019) (“[A] true threat does not need to be communicated directly to the intended victim.”); Porter v. Ascension Parish Sch. Bd., 393 F.3d 608, 616 (5th Cir. 2004) (“[To] be lawfully punished, the threat must be intentionally or knowingly communicated to either the object of the threat or a third person.”) (emphasis omitted)); Taupier, 193 A.3d at 30-31 & n.27 (collecting cases). 3 Finally, the harms associated with criminal threats are severe enough to overcome otherwise applicable policies, even those of constitutional dimension. See Counterman v. Colorado, 600 U.S. 66, 74 (2023) (explaining that true threats of violence constitute a “historically unprotected category of communications” where the First Amendment is concerned). 3 The Supreme Court of Connecticut in Taupier justified the criminalization of indirect threats by reasoning that a serious threat issued only to third parties will often be communicated to its target. Id. at 32. We respectfully disagree with Taupier, however, to the extent it suggests that threats are condemnable based only on the harm they cause to their targets. Cf. id. at 33 (affirming a disorderly-conduct conviction based on the harm the defendant’s threat caused to a third party who heard it). 18 C. The Threats Exception With these preliminary principles in mind, we turn to the question before us: whether the attorney-client privilege shields threats like Mr. Moore’s. Rather than parsing whether Mr. Moore’s statements meet the technical elements of the attorney-client privilege, we hold that even if the threats met the privilege’s ordinary requirements, the privilege would still not apply for a more fundamental reason. The attorney-client privilege does not protect communications that themselves constitute criminal threats to cause death or substantial bodily harm. We reach this conclusion because exempting serious, objectively credible threats from the attorney-client privilege (1) prevents clients from abusing the privilege (and therefore flows logically from the foundation for the crime/fraud exception to the privilege) and (2) is grounded in strong historical roots. 1. Abuse of the privilege We begin at common ground. Our dissenting colleagues recognize that the attorney-client privilege does “not apply where the attorney-client relationship is being ‘misused’ or ‘abused’ to advance criminal ends.” Post dissent at 73 (quoting In re Pub. Def. Serv., 831 A.2d at 909-10). We concur. The difference between our view and that of the dissent, however, flows from how we define “abuse” and “advanc[ing] criminal ends.” In our view, uttering a serious, criminal threat in the 19 presence of one’s attorney constitutes an abuse of the attorney-client relationship and therefore precludes application of the attorney-client privilege to the communication that constitutes the threat. In the dissent’s view, it does not. We explain below why we think our position is the correct one. Our reasoning is similar 4 to that underlying the crime/fraud exception to the attorney-client privilege. One of the ways in which we have defined this exception is as follows: the “attorney-client privilege does not apply to a communication occurring when a client, . . . regardless of the client’s purpose at the time of consultation [with her lawyer], uses the lawyer’s advice or other services to engage in or assist a crime or fraud.” In re Pub. Def. Serv., 831 A.2d at 908 (quoting Restatement (Third) of the Law Governing Lawyers § 82 (2000)). This exception, as we alluded to above, rests on the idea that a client “abuses” her relationship with her attorney by using the relationship to commit an unlawful act. See id. at 908. To allow a client who utters a criminal threat in the presence of her attorney to avail herself of the attorney-client privilege would be to countenance just such an abuse. Outside of the context of her relationship with her attorney, the client could 4 Mr. Moore and the dissent argue that this jurisdiction’s crime/fraud exception does not apply to his threats, primarily contending that the exception is irrelevant unless the attorney’s services further the commission of a crime. But because, rather than rely on the crime/fraud exception, we craft a new exception for criminal threats, we need not address the crime/fraud exception’s scope here. 20 not utter criminal threats in an effort to intimidate those around her; if frustrated with a delay in her Starbucks order, for instance, the client could not in the presence of another customer threaten to kill the barista. But if the client were frustrated with a delay in her legal proceeding, the client could threaten to kill someone in the courthouse—safe in the knowledge that her attorney could not be called to testify about the threat—and subject her attorney to the discomfort, fear, or outrage 5 that may stem from hearing objectively credible threats. We would call such a course of action an abuse of the attorney-client relationship—our hypothetical client uses a privilege designed to strengthen the relationship between her and her attorney to enable her to commit a criminal act that will likely harm her relationship with her attorney. That is an abuse. 6 See Edward J. Imwinkelried, supra, § 6.13.2(d) (“The 5 Mr. Harvey’s response to Mr. Moore’s threats exemplifies this point—he was disturbed by Mr. Moore’s threats. After the first threat, he indicated he did “not want to participate” in Mr. Moore’s plans and “labored over” the dilemma Mr. Moore had placed him in. The seriousness with which he approached Mr. Moore’s first threat is evident from how he spoke with Mr. Moore afterward—he said, “You will never, ever use this kind of language with me about anybody . . . .” And when Mr. Moore threatened the AAG for a second time, Mr. Harvey was “so concerned”—indeed, “extremely concerned”—that he interrupted a hearing to ask the presiding judge to recall Mr. Moore’s case. 6 In response to this concern, the dissent tells attorneys to toughen up, suggesting that to be discomfited by hearing a criminal threat makes one a “pearl- clutching milquetoast[ ].” Post dissent at 102 & n.28. We, however, are not persuaded that an attorney—as a condition of engaging in client representation— must assent to being at least a witness to a crime. Nor are we convinced by the 21 law attempts to remove disincentives to lawful consultations ‘but not at the price of encouraging illegal conduct.’” (quoting Am. L. Inst., Model Code of Evidence 162 (1942))). We need not and do not rest this exception solely on harm to the attorney and her relationship with her client. Allow us to tweak the above hypothetical. The client now has a hard-boiled attorney who is not intimidated by her client’s threat to a third party. But the attorney feels morally obligated to report the threat to law enforcement, who in turn warn the threat’s target. And the target, not as accustomed to danger as is the hard-boiled attorney, is terrified. Here again we have an abuse. The client in this hypothetical has sowed exactly the harm at the core of the prohibition against criminal threats—she has caused the target of her threat to fear dissent’s suggestion that because an attorney may hear a client describe a past crime, she will not suffer if she is present when her client commits a crime. We recognize, of course, that the Ninth Circuit has embraced similar logic to that of the dissent in the context of the psychotherapist-patient privilege. Sitting en banc, that court held that threats against federal officers communicated to the defendant’s psychiatrist were privileged because “[o]nce [he] finished uttering the threats, the charged crime was completed, and the psychiatrist was in the same position she would have occupied had her patient described a bank robbery in which he had participated a week earlier.” United States v. Chase, 340 F.3d 978, 982 & n.1 (9th Cir. 2003) (en banc). Our response to the Ninth Circuit is similar to our response to the dissent—we respectfully disagree that hearing one’s client issue a threat is the same as listening to one’s client describe a past bank robbery. It is closer, we think, to watching one’s client rob the bank. 22 for their life. 7 See Black, 538 U.S. at 360. To allow a client to sow fear, through their attorney and free of consequence, is to allow an abuse of the attorney-client privilege. 8 See 1 Kenneth S. Broun et al., McCormick on Evidence § 95 (8th ed. 2022) (“The privileged communications may be a shield of defense as to crimes already committed, but it cannot be used as a sword or weapon of offense to enable 7 As the District observes, clients threatening those involved in their case (such as attorneys) is not a mere theoretical concern. See, e.g., Haney v. United States, 41 A.3d 1227, 1229 (D.C. 2012) (threatening a testifying police officer); In re Grand Jury Investigation, 902 N.E.2d 929, 930 (Mass. 2009) (communicating to counsel a threat to kill the judge); State v. Perkins, 626 N.W.2d 762, 765 (Wis. 2001) (threatening a judge); United States v. Tanner, 26 F. App’x 469, 470, 472 (6th Cir. 2001) (per curiam) (nonprecedential) (threatening to harm counsel). And trial participants have good reason to fear such threats. See, e.g., Jenkins v. United States, 80 A.3d 978, 996-97 (D.C. 2013) (upholding a trial court’s finding that the defendants were involved in the shooting of a potential witness in order to prevent his testimony); Ward v. United States, 55 A.3d 840, 849 (D.C. 2012) (concluding that sufficient evidence supported the trial court’s finding that the defendant hired his codefendant to kill an adverse witness); Roberson v. United States, 961 A.2d 1092, 1096-97 (D.C. 2008) (upholding a trial court’s finding that the defendant conspired to murder a witness to prevent him from cooperating with the prosecution). 8 The dissent suggests that the above circumstance cannot constitute an abuse because one who threatens someone in the presence of their attorney likely does not intend for their attorney to pass the threat on to law enforcement. But in the now-reinstated portion of the division opinion in this case, this court held that one who utters a threat may be convicted without proof that she intended the threat to reach its target. Moore, 285 A.3d at 236. We question, therefore, why the client’s intent with respect to the dissemination of their threat bears on the abuse question. Instead, what makes the above circumstance an abuse is the effect of the (likely) disclosure on the threat’s target. 23 persons to carry out contemplated crimes against society.” (quoting Gebhardt v. United Rys. Co. of St. Louis, 220 S.W. 677, 679 (Mo. 1920))). Although our dissenting colleagues vigorously contest our above reasoning, they are unwilling to commit to complete application of the attorney-client privilege to criminal threats; they are silent as to whether “a threat of death or serious bodily injury against the lawyer themself could constitute an abuse of the relationship such that the privilege should not apply.” Post dissent at 104 n.30. For two reasons, we see no merit in such a distinction. First, objectively credible threats can inspire fear, intimidation, or other strong emotion in the listening party even if the threatened violence will fall on someone else. See Ivers, 967 F.3d at 716. Second, and relatedly, our criminal-threat statutes do not make distinctions based on a threat’s target but instead punish both threats targeted at third parties and threats targeted at the listener. See Carrell, 165 A.3d at 320. This is so in part because—as was the case here—third parties who hear objectively credible threats are often likely to pass them on to their targets. See Taupier, 193 A.3d at 32. We also find the dissent’s silence regarding attorney-target threats telling. In our view, the dissent’s reluctance to commit reflects its belief that attorney-targeted threats are more harmful than are threats directed at third parties. If we are right about that, then the dispute between the majority and the dissent is grounded not in 24 principle, but instead in line-drawing—how much harm is enough to constitute an abuse? For the reasons stated above, we are confident in where we have drawn that line. Because making a serious, criminal threat in the presence of one’s attorney constitutes an abuse of the attorney-client relationship, the attorney-client privilege does not protect the communication constituting the threat. 2. Historical Support Treating criminal threats uttered to an attorney as unprivileged not only fits well with the reasoning underlying the crime/fraud exception but also finds strong historical footing. In the first half of the nineteenth century, a widely read New York case explained that when a client “is not disclosing what has already happened, but is actually engaged in committing the wrong, he can have no privileged witness.” Coveney v. Tannahill, 1 Hill 33, 41 (N.Y. Sup. Ct. 1841) (citing Greenough v. Gaskell (1833) 39 Eng. Rep. 618; 1 MY. & K. 98); see also 8 Wigmore, supra, § 2298 (quoting Coveney at length). Since then, many American courts have excluded from the attorney-client privilege the commission of crimes completed in the presence of the attorney. 9 9 See State v. Mewherter, 46 Iowa 88, 94 (1877) (concluding that the client’s threats were unprivileged because they “in no manner pertained to the business of the professional consultation” but declining to address whether “threats[,] in the 25 To be sure, the cases from other jurisdictions do not all mirror the rule we adopt today, nor need we endorse their particular holdings. The point remains, however, that in each case the court either (a) suggested that crimes—or at least threats—completed in the presence of the attorney were not privileged or view of their unlawfulness,” were categorically deprived “of the character of a privileged communication”); McMannus v. State, 39 Tenn. 213, 216 (1858) (“It would be monstrous to hold, that if counsel was asked and obtained, in reference to a contemplated crime, that the lips of the attorney would be sealed, when the fact might become important to the ends of justice in the prosecution of crime. In such a case[,] the relation cannot be taken to exist. Public policy would forbid it. We presume the rule has never been extended so far, nor will it be.”); Hopkinson v. State, 632 P.2d 79, 116 (Wyo. 1981) (“We cannot imagine a threat of injury made by a client toward the family and property of an attorney as being privileged and within those communications protected.”); State v. Hansen, 862 P.2d 117, 121 (Wash. 1993) (“If an attorney-client relationship could have been found to exist when Hansen made the threat against the judge, the prosecutor, and the public defender, the privilege would still not apply. The attorney-client privilege is not applicable to a client’s remarks concerning the furtherance of a crime, fraud, or to conversations regarding the contemplation of a future crime.”); United States v. Alexander, 287 F.3d 811, 816 (9th Cir. 2002) (“Alexander’s threats to commit violent acts against [his counsel] and others were clearly not communications in order to obtain legal advice.”), abrogated on other grounds by United States v. Plouffe, 445 F.3d 1126, 1129 (9th Cir. 2006); United States v. Ivers, 967 F.3d 709, 716 (8th Cir. 2020) (“Threats of violence are not statements that fall under the scope of the attorney-client privilege.”); see also People v. Dang, 113 Cal. Rptr. 2d 763, 766-67 (Ct. App. 2001) (holding that a California statutory evidence rule prohibits application of the attorney-client privilege to a client’s threat to kill a witness); semble State v. Nixon, 207 P. 854, 855 (Kan. 1922) (quoting, with apparent approval, the trial court’s hypothetical that “I don’t think anybody would claim that, if the defendant had gone into [the attorney’s] office, and said to him, ‘I am going to kill [the victim],’ and then consulted with him, it would claim such a matter was privileged”); Jackson v. State, 293 S.W. 539, 540 (Tenn. 1927) (“[T]hreats made by a client against the life of a person during a professional consultation with his attorney are not privileged.” (quoting 28 Ruling Case Law 559, 560)). 26 (b) concluded that the specific threat or crime before it was not privileged. We thus break little new ground in holding today that serious threats issued to an attorney fall outside the attorney-client privilege. We are aware of only two contrary precedents, and neither persuades us. See In re Grand Jury Investigation, 902 N.E.2d 929 (Mass. 2009); State v. Boatwright, 401 P.3d 657 (Kan. Ct. App. 2017). 10 In In re Grand Jury Investigation, a case 10 Mr. Moore directs us to two slightly older cases as well: Purcell v. District Attorney, 676 N.E.2d 436 (Mass. 1997), and Newman v. State, 863 A.2d 321 (Md. 2004). We do not see our holding as in direct conflict with either case because neither case involved a criminal threat, and Purcell did not even involve a completed crime. Purcell concerned alleged attempted arson, and the court concluded only that “[a] statement of an intention to commit a crime made in the course of seeking legal advice is protected by the privilege.” Purcell, 676 N.E.2d at 438, 441. Newman, on the other hand, involved the agreement portion of a conspiracy, which is sufficient to complete the crime of conspiracy under Maryland law (even if it would not suffice under the District of Columbia’s general conspiracy statute). Newman, 863 A.2d at 324. Compare Belton v. United States, 581 A.2d 1205, 1209 (D.C. 1990) (requiring, beyond an agreement to commit a crime, “at least one overt act . . . in furtherance of the common scheme”), with Townes v. State, 548 A.2d 832, 834 (Md. 1988) (“In Maryland, the crime [of conspiracy] is complete when the unlawful agreement is reached, and no overt act in furtherance of the agreement need be shown.”). Nevertheless, Newman did not concern a completed criminal threat, and the court in Newman addressed only the crime/fraud exception and held that a mere statement of intent to commit a crime was insufficient for application of the exception—a question we do not reach today. Newman, 863 A.2d at 335 (“We . . . join our colleagues on both the federal and state levels who have required more than a mere statement of the intent to commit a crime or fraud to trigger the crime-fraud exception to the attorney-client privilege.”). Neither the Purcell nor the Newman court had occasion to opine on whether the privilege would apply specifically to the crime of criminal threats. For reasons discussed above, we see this distinction as significant. 27 involving a client’s alleged threat to harm a judge and her family, the Massachusetts Supreme Judicial Court held “that a client’s communications to his lawyer threatening harm are privileged unless the crime-fraud exception applies.” 902 N.E.2d at 930, 934. The court explained that “[i]f a lawyer suspects that the client intends to act on an expressed intent to commit a crime, the lawyer may attempt to dissuade the client from such action, and failing that, may make a limited disclosure to protect the likely targets.” Id. at 933. The Kansas Court of Appeals reached a similar result. It held that a client’s criminal threat to kill his ex-fiancée fell within the attorney-client privilege because the alternative approach might disincentivize clients from communicating threats to the attorney and discourage attorneys from warning the targets. See Boatwright, 401 P.3d at 660-61, 664. Mr. Moore urges us to follow suit. The Massachusetts court justified its holding primarily by reference to a previous case, in which it held that “a statement of an intention to commit a crime made in the course of seeking legal advice is protected by the privilege, unless the The dissent finds these distinguishing factors unpersuasive because, in its view, there is no meaningful difference between the statements in Purcell and Newman and completed threats. For the reasons articulated above, we do not see our holding as conflicting with Purcell and Newman. But if those cases indeed stand for the proposition that the attorney-client privilege covers completed criminal threats of death or serious bodily injury, then we find them unpersuasive for the same reasons we are unpersuaded by In re Grand Jury Investigation and Boatwright. 28 crime-fraud exception applies.” In re Grand Jury Investigation, 902 N.E.2d at 932-33 (internal quotation marks omitted). It saw “no reason to depart” from that decision in the context of threats. Id. at 933. But we see a great deal of difference between a mere statement of intention to commit a crime and a criminal threat of serious bodily harm or death—as we explained above, serious, objectively credible criminal threats sow fear and intimidation in a way that other statements of intention may not. This difference in degree of harm justifies different treatment where the attorney-client privilege is concerned. The Kansas court’s justification fares no better—although it suggests that permitting attorneys to disclose their clients’ threats to authorities or the targets would remedy the harm posed (at least to the targeted party) by those threats, its solution solves little. See Boatwright, 401 P.3d at 664. First, the Kansas court does not account for the harm that the attorney may suffer due to hearing the threat. Second, as we indicated above, we do not agree that disclosure alone protects the target. Archetypically, speakers issue their threats directly to the target, meaning the target learns of the threat immediately. It would be strange to conclude, then, that, in cases where the threat is conveyed to a third party, the criminal harm abates merely because the target later learned of the threat. That simply puts the recipient in the same position as if the threat had been issued to them directly. We are thus unpersuaded by the minority approach of Massachusetts and Kansas. 29 Our dissenting colleagues reject our appeal to history because no jurisdiction has yet adopted a free-standing threats exception. But by focusing their critique on the novelty of such an exception, our dissenting colleagues miss the forest for the trees. Although we may stand alone in our specific reasoning, the cases we cite above demonstrate that many other jurisdictions withhold the protection of the attorney-client privilege from criminal threats one way or another. It is the dissent’s result—and not ours—that is out of step with the majority approach. 11 As for our reasoning, we elect a freestanding threat exception over, for instance, holding that the crime/fraud exception encompasses threats, to provide more protection for the attorney-client relationship than that provided by the approaches of other jurisdictions. See post DEAHL concurrence at 58 (arguing that the crime/fraud exception applies to criminal threats and therefore that all criminal threats—not merely threats of serious bodily harm or death—fall outside of the privilege). Put differently, the path we take is motivated by the same concerns about chilling attorney-client relations that the dissent expresses. We do not see our efforts at compromise as betraying a weakness in our position. 11 Indeed, what one considers unprecedented often depends on perspective. Were we to adopt Mr. Moore’s and the dissent’s view, we would in effect be creating an attorney-client exception to the criminal threats statute. We view that outcome as unprecedented. 30 * * * Although we conclude that Mr. Moore’s threats were not privileged, we need not—and do not—decide whether all completed crimes, or even all completed criminal threats, fall outside the attorney-client privilege. Because Mr. Moore threatened to shoot the AAG, it suffices today to hold that the privilege excludes completed criminal threats of death or serious bodily injury. 12 Cf. Perez Hernandez, 286 A.3d at 995 (“We focus instead, in the common law tradition, on deciding the case before us.”); Hedgepeth v. Whitman Walker Clinic, 22 A.3d 789, 812 (D.C. 2011) (en banc) (“The development of common law proceeds on a case-by-case incremental basis, and it is on that solid factual ground that judicial opinions build on a framework for analysis based on certain general precepts.”). 13 12 By leaving the treatment of lesser threats unsettled, we by no means indicate support for the eventual elimination of the privilege with respect to those statements. Should this court in the future confront the question of whether the privilege should remain in force with respect to lesser threats, it is our view that this opinion does not dictate the outcome of that case. 13 In the District, the attorney-client privilege remains a judicial creation. See In re Ti.B., 762 A.2d 20, 27-28 (D.C. 2000). We therefore do not treat the elements of the attorney-client privilege, or Wigmore’s formulation of the privilege, as if they were “the words of [a] statute.” Gallimore v. Washington, 666 A.2d 1200, 1208 (D.C. 1995) (distinguishing between common-law analysis and statutory interpretation). Instead, we look to how our court and other courts have interpreted the privilege over the years and the reasons underlying the privilege to guide our analysis. And we remain cognizant that the common law is “capable of growth and 31 We confine our present decision to threats of death or serious bodily injury both out of respect for the attorney-client privilege and in recognition of the concern that we and other jurisdictions have exhibited for crimes posing a grave risk to life or limb. It may be that exposing attorney-client communications that constitute other crimes or lesser threats poses greater risk to client candor and the integrity of the adversarial process than does revealing the serious threats we confront today. Cf. Counterman, 600 U.S. at 75, 80 (concluding that some true threats must be protected in order to prevent too strong of a chilling effect on protected speech, but leaving reckless, knowing, and intentional threats unprotected so as to appropriately balance this chilling effect with the “profound harms, to both individuals and society, that attend true threats of violence”). Moreover, we and other jurisdictions have approved rules that distinguish communications regarding “a criminal act that the lawyer reasonably believes is likely to result in death or substantial bodily harm absent disclosure” as less worthy of an attorney’s confidentiality. 14 D.C. R. Pro. development at the hands of judges.” Id. (quoting Nelson v. Nelson, 548 A.2d 109, 112 (D.C. 1988)). After all, that is what distinguishes the common law from statutes. 14 We do not address whether Rule 1.6(c)(1) by its own force sheds the confidentiality required for the attorney-client privilege in the first place. Compare Adams v. Franklin, 924 A.2d 993, 999-1000 (D.C. 2007) (“The attorney-client privilege protects only communications from a client to an attorney that are, at the time they are communicated, intended to be confidential.”), with id. at 999 n.6 (distinguishing the applicability of Rule 1.6 from the issue of whether a communication was confidential for purposes of the attorney-client privilege), and D.C. R. Pro. Conduct 1.6 cmt. 6 (“This rule is not intended to govern or affect 32 Conduct 1.6(c)(1); accord Md. R. Pro. Conduct 1.6(b)(1); Va. R. Pro. Conduct 1.6(b)(7); Model R. Pro. Conduct 1.6(b)(1). At least one legislature has codified such a rule for its attorney-client privilege. See Cal. Evidence Code § 956.5 (West 2005) (“There is no privilege under this article if the lawyer reasonably believes that disclosure . . . is necessary to prevent a criminal act that the lawyer reasonably believes is likely to result in the death of, or substantial bodily harm to, an individual.”); Dang, 113 Cal. Rptr. 2d at 767 (applying Section 956.5 to a criminal threat). We thus leave for another day whether to exempt other crimes and lesser threats from the attorney-client privilege. 15 judicial application of the attorney-client privilege . . . .”). We observe only that the rationale for differentiating communications risking death or serious bodily harm for purposes of Rule 1.6’s confidentiality requirement may justify a similar distinction for purposes of the attorney-client privilege. 15 The presence of an objectively credible, serious threat in an otherwise-privileged attorney-client conversation does not necessarily render the entire conversation admissible. Trial courts are well-positioned to separate the privileged portions of communications from the unprivileged. Cf. In re Grand Jury Subpoena, 419 F.3d 329, 343 (5th Cir. 2005) (“[T]he proper reach of the crime-fraud exception when applicable does not extend to all communications made in the course of the attorney-client relationship, but rather is limited to those communications and documents in furtherance of the contemplated or ongoing criminal or fraudulent conduct.”); In re Vargas, 723 F.2d 1461, 1467 (10th Cir. 1983) (“[T]he trial court need only conduct an in camera inspection of the documents if there is a possibility that some of them may fall outside the scope of the [crime/fraud] exception to the privilege.”); In re Sealed Case, 676 F.2d 793, 812 n.74 (D.C. Cir. 1982) (“Once a sufficient showing of crime or fraud has been made, the privilege vanishes as to all material related to the ongoing violation.” (emphasis added)). Before the en banc court and the division, Mr. Moore did not present a fallback argument that even if 33 The dissent criticizes us for not limiting our threats exception still further, pointing out that we have not required as a prerequisite to the exception’s application disclosure of the threat by counsel, proof beyond a reasonable doubt that the communication constituted a threat, use only in criminal cases, or some likelihood that the threat be carried out. Each of the above criticisms, however, could be levied against the already extant crime/fraud exception. See In re Pub. Def. Serv., 831 A.2d at 902 (imposing none of the above requirements when setting out the crime/fraud exception’s contours). We see no reason to distinguish our newly articulated threats exception from the crime/fraud exception by adding additional limitations or procedures not present there. D. Counterarguments Mr. Moore and amicus the Public Defender Service (PDS) warn against withholding the attorney-client privilege from Mr. Moore’s statements for two reasons. In their view, such an outcome will (1) unduly chill attorney-client candor evidence of his actual threatening statements were properly admitted, the trial court erred by admitting more than the threats themselves. As a result, Mr. Moore also did not present any argument about exactly what statements should have been excluded on such a theory and did not address whether the admission of such statements would have been harmful error. Because any such argument has been abandoned on appeal, see Abdus-Price v. United States, 873 A.2d 326, 332 n.7 (D.C. 2005) (deeming an argument not raised on appeal “to be abandoned”), we do not address those issues and express no view on them. 34 and (2) undermine public safety. And the dissent to a degree shares their concerns. Although we acknowledge that our decision involves tradeoffs, we are unpersuaded that the above concerns justify a different result. 1. Chilling Attorney-Client Candor Mr. Moore, PDS, and the dissent argue that statements like Mr. Moore’s threats are vital to the legal relationship because they express the client’s frustration with the legal proceeding. Excluding such expressions from the attorney-client privilege, in their view, would impinge on attorney-client candor and require clients to censor themselves when speaking to their attorneys. Specifically, PDS warns that even if threatening statements themselves are not worth protecting with the privilege, exempting threats from the privilege could chill other nonthreatening speech. PDS fears that clients may “swallow words that are in fact not true threats” in order to give “the unlawful zone” a wide berth. Counterman, 600 U.S. at 78 (quoting Speiser v. Randall, 357 U.S. 513, 526 (1958)). We are unpersuaded by these arguments for three reasons. First, clients are presumably used to communicating without pronouncing objectively credible threats because the prohibition on issuing criminal threats applies to our day-to-day 35 conversations as well. This makes the exception to the privilege for serious criminal threats unlikely to impede clients’ ability to communicate with their attorneys. 16 Second, to the extent that excluding criminal threats from the privilege does impede some client candor, this is hardly a new development. As previously discussed, most courts over the years have excluded criminal threats from the privilege, and we have never before suggested that criminal threats would be privileged (at least until the now-vacated panel decision in this case). Moreover, the privilege already operates with other, similar exceptions, such as the crime/fraud exception. Clients in our jurisdiction and others must moderate their conversations with their attorney regardless of the outcome of this case. And third, because attorneys can already disclose serious, objectively credible threats made by their clients, the marginal chilling effect of allowing attorneys to testify is minimal. See Chase, 340 F.3d at 996-97 (Kleinfeld, J., concurring in the 16 Our dissenting colleagues respond that because individuals consult with lawyers to respond to stressful circumstances, they deserve “a bit more breathing room” in their conversations with lawyers. Post dissent at 126. We agree, which is why we hold only that criminal threats of death or serious bodily harm are excluded from the privilege. Moreover, as we state elsewhere in our opinion, see infra at 39-40, we trust that attorneys can and will distinguish between emotional, not objectively credible statements and objectively credible, criminal threats when deciding whether to disclose a client’s statement. So, we believe that the exception we announce today leaves enough room for clients to express their emotions to their lawyers. 36 result). The dissent points out that retaining the attorney-client privilege, in addition to facilitating the administration of justice, safeguards the client’s privacy and protects the interpersonal relationship between the attorney and their client. But the dissent concedes that confidentiality rules allow attorneys to disclose serious threats of bodily injury or death made by their clients. We do not see what privacy interest a client can claim in information (a threat) all agree an attorney can disclose to third parties. And once an attorney breaks confidentiality by disclosing the threat, we doubt much if any of the relationship between the attorney and the client will survive. Indeed, this case proves that very point. Once Mr. Harvey disclosed Mr. Moore’s threats to the court—and before any prosecution related to the threats had begun— Mr. Harvey withdrew from representing Mr. Moore because his relationship with Mr. Moore had deteriorated to the point where further representation was impossible. Privileging criminal threats, moreover—as Mr. Moore and PDS would have us do—also risks sacrificing the attorney-client relationship. Consider, as Judge Thompson did in her division dissent, if Mr. Moore had threatened Mr. Harvey’s family rather than the AAG as a means of venting his frustration: “You are doing a terrible job for me. I know where you and your family live, and I am going to torture and kill you all.” Moore, 285 A.3d at 259 (Thompson, J., dissenting). Such a statement involves communications at least as revealing of the 37 client’s frustrations as Mr. Moore’s threats in this case. As a result, they would have as strong a claim to privilege under the rule Mr. Moore endorses. Under Mr. Moore’s interpretation of the privilege (and, perhaps, the dissent’s), the attorney would be left without the ordinary protections of arrest and criminal prosecution against the threat of serious violence. Such a burden would not strengthen the attorney-client relationship or better facilitate the adversarial process—quite the opposite. 2. Public Safety Setting aside clients’ interest in privileged communications, Mr. Moore and PDS caution that our holding undermines public safety. They argue that unless criminal threats are kept privileged, (1) clients will not issue the threats before their attorneys, and consequently the attorneys will not be able to dissuade them from harming others; and (2) when clients do convey threats to their attorneys, the attorneys will not disclose them to law enforcement or the target. And, although the dissent does not embrace the above argument, it suggests that (3) our exception “is not tailored to any real-time or even future public safety objective.” Post dissent at 122. We doubt all three propositions. With respect to the first, we think it unlikely that many defendants will change their behavior because of our holding. Mr. Moore and PDS appear to imagine a 38 defendant who (1) wishes to threaten, in the presence of their attorney, someone with death or serious bodily harm and (2) takes into consideration whether or not such a communication would be privileged. Yet a client so well-versed in our attorney-client privilege doctrine presumably also knows that, regardless of which rule we adopt today, their attorney may disclose the threat to the police and the target under Rule 1.6. If the client knows their attorney might disclose their threat to the police, we doubt they would utter the threat regardless of whether the privilege applies—unless, of course, they are hoping their attorney will convey the threat to the target for them. At any rate, whatever benefit the attorney-client privilege would provide to public safety by encouraging clients to divulge their threats to counsel, the privilege’s protection would be quite underinclusive because it would not apply to most client threats. Most threats—even those uttered in the presence of an attorney—are presumably not in pursuit of legal advice and thus lack privilege either way. Those unprivileged threats are just as dangerous as threats seeking legal advice, which, under Mr. Moore’s rule, would be privileged. If the privilege is supposed to protect us from client threats, it does a poor job because it does not even apply to most client threats. 39 Nor are we convinced by the argument that we must apply the attorney-client privilege to criminal threats to allow attorneys to dissuade their clients from acting out those threats. When a client consults a lawyer as to whether they might ask a witness not to testify, no crime has yet been committed. But when a client issues a serious, objectively credible, criminal threat to a witness in the presence of their attorney, the crime is complete; the attorney cannot through sober, course-correcting advice put the genie back in the bottle. Whatever part the lawyer might play in persuading their client not to commit the additional crime of following through on their threat, the crime of a criminal threat has already been committed, 17 and the lawyer at least has already suffered the harms associated with that crime. To justify those harms based on the mere possibility that the lawyer might prevent, through the power of reasoning, a future crime both underestimates the harm that threats cause and overestimates attorneys’ powers of persuasion. As for whether attorneys will change their behavior in response to our holding, we doubt this too. Attorneys will not necessarily disclose all threatening statements made by their client in their presence. And for good reason. Facially threatening 17 Perhaps an attorney’s immediately subsequent conversation with their client could reveal that the threat was not objectively credible. As we have discussed above, whether a threat is objectively credible and thus criminal turns in part on the surrounding context, including the other statements made by the issuer. Such a situation is not before us, and we need not address it. 40 statements a client makes in the presence of an attorney may well be idle, not objectively credible, and thus not crimes. As the division correctly observed, “when a threat is delivered only to a third party, that fact can bear on” “one’s ‘mens rea to utter the words as a threat’” or whether the threat is objectively credible. Moore, 285 A.3d at 236 n.9 (quoting Carrell, 165 A.3d at 317). We emphasize that the exception we announce today applies only when a judge preliminarily determines that a criminal threat crime (one involving serious bodily injury or death) has occurred. See In re Pub. Def. Serv., 831 A.2d at 902. Most statements, when viewed in context, will not qualify as such. But if an attorney believes that the above standard is met, we believe that that attorney—an officer of the court—would report the threat under Rule of Professional Conduct 1.6(c) irrespective of whether the privilege applies. To be sure, their disclosure now carries a greater risk of subjecting their client to investigation and prosecution. But even if the attorney could not testify, the client would still likely face investigation and possible prosecution after the attorney reported the threat. So, the difference in attorney incentives is minimal. See Newman, 863 A.2d at 344 (Wilner, J., dissenting) (“I cannot conceive, and the Court offers no explanation, of why a lawyer who believes that a disclosure is necessary to prevent death or serious bodily harm to another will feel free to make a disclosure under Rule 1.6, knowing that, as a result, his client will almost certainly be the target of a criminal 41 investigation, but will nonetheless be reluctant to make the disclosure because he/she may be called to testify in court.”). Finally, the dissent’s skepticism as to whether a threats exception will improve public safety is grounded in too rosy a view of the power of disclosure alone to prevent disruption of the target’s life in response to the threat, protect the target in the event that the threatener attempts to carry out the threat, and disincentivize future threats. Law enforcement cannot serve as the targets’ personal bodyguards, and most individuals cannot afford their own private security, let alone security for their family members. The fact that law enforcement is “either investigating or pursuing charges” with respect to the threat, post dissent at 107, is likely to be of cold comfort to a threat’s target; we struggle to see how the government could obtain probable cause to arrest—let alone persuade a jury that a threat occurred—if the threat itself cannot be admitted into evidence, regardless of how dogged an investigation is conducted. 18 And because in the absence of prosecution the issuer 18 To draw this issue into starker relief, consider the following scenario. An individual previously convicted of assaulting a victim is released from prison and charged with once again assaulting the same victim. He is released pending trial, and he tells his attorney that he plans to kill the complainant unless she agrees not to testify against him. Even if defense counsel disclosed the threat to law enforcement, under Mr. Moore’s position the disclosed information remains privileged and the court would be powerless to use it as a basis to revoke the individual’s release or impose a stay-away order. Notifying law enforcement, then, is hardly an enduring solution when the only evidence of the threat is excluded in legal proceedings. 42 of the threat would face no legal consequences, that issuer would have no incentive to refrain from issuing future threats in the same manner. Even in the absence of an attempt by the threat’s issuer to follow through on their threat, threats, once disclosed, harm their targets. By disincentivizing threats, the exception we announce prevents those harms and thereby protects public safety. III. Conclusion For the foregoing reasons, we affirm the Superior Court’s conclusion that the attorney-client privilege does not apply to Mr. Harvey’s testimony. We return the appeal to the division to address Mr. Moore’s remaining arguments. So ordered. MCLEESE, Associate Judge, concurring: The opinion for the en banc court holds that Mr. Moore’s threats to shoot the prosecutor are not protected by the attorney-client privilege. I agree with that holding, and I join the opinion for the en banc court in full. I write separately to briefly explain an additional basis for that holding. The attorney-client privilege protects only communications “made in confidence.” Jones v. United States, 828 A.2d 169, 175 (D.C. 2003) (internal quotation marks omitted). In determining whether a communication between a 43 client and an attorney was made in confidence, we consider what the client reasonably believed. Id. at 176. In my view, the record in this case amply supports a conclusion that Mr. Moore could not reasonably have believed that his threats to shoot the prosecutor would be kept confidential. Where applicable, the attorney-client privilege prevents “compelled disclosure of privileged communications.” Adams v. Franklin, 924 A.2d 993, 998 (D.C. 2007). Even where disclosure is not being compelled, however, attorneys have a general fiduciary duty of confidentiality to their clients. See generally, e.g., Herbin v. Hoeffel, 806 A.2d 186, 197 (D.C. 2002) (“[D]isclosure of client confidences is contrary to the fundamental principle that the attorney owes a fiduciary duty to [the attorney’s] client.”) (internal quotation marks omitted). Thus, attorneys can be held liable for wrongfully disclosing a client’s confidential information. See, e.g., id. at 197-98 (remanding for further consideration of claim that attorney committed tort of intentional infliction of emotional distress by impermissibly disclosing client confidences); see generally Restatement (Third) of the Law Governing Lawyers §§ 16(3) (Am. L. Inst. 2000) (lawyer has duty to “comply with obligations concerning the client’s confidences”), 49 (lawyer is civilly liable to client if lawyer breaches fiduciary duty concerning client confidences). 44 A lawyer’s fiduciary duty of confidentiality is not absolute. See generally Restatement (Third) of the Law Governing Lawyers §§ 61-67 (listing exceptions to general fiduciary duty of confidentiality). One exception to the duty of confidentiality recognized by the Restatement seems highly relevant to this case: “A lawyer may use or disclose confidential client information when the lawyer reasonably believes that its use or disclosure is necessary to prevent reasonably certain death or serious bodily harm to a person.” Id. § 66(1); see also id. cmt. c (“[T]he character of a threatened act as a crime . . . suggests a state of mind on the part of the perpetrator that is more threatening to the intended victim than an act that is merely negligent or not wrongful, and thus may more readily warrant the lawyer’s conclusion that the risk of harm is great.”); cmt. d (“[A] lawyer is permitted to act under [this] Section whenever the lawyer reasonably believes that intervention is necessary in response to a present threat, even if the death or serious bodily harm appears likely to occur some time in the future.”). For purposes of this appeal, the following facts appear to be undisputed. During a break before the beginning of his criminal trial, Mr. Moore became very agitated and angrily made statements to his defense counsel threatening to shoot the prosecutor. When defense counsel expressed concern that Mr. Moore was serious, Mr. Moore said “goddamn right” and repeated his threat to shoot the prosecutor, adding that he owned guns. Later that day, Mr. Moore told defense counsel that he 45 did not mean to act on his threats. Defense counsel told Mr. Moore that defense counsel would believe Mr. Moore if Mr. Moore repeated his threats. Defense counsel also told Mr. Moore that defense counsel would inform the court if Mr. Moore made similar threats in the future. The trial was continued for several months. During another break after trial was resumed, Mr. Moore was “out of control,” extremely angry, and agitated. Mr. Moore said twice that if he lost his job, he was “going to bust a cap in this bitch.” Mr. Moore also took his hand and made a gesture as though repeatedly shooting a gun. Defense counsel understood Mr. Moore to be threatening to shoot the prosecutor. Defense counsel was extremely concerned and believed that Mr. Moore meant what Mr. Moore was saying. Defense counsel then disclosed Mr. Moore’s threats to the court. This court has not yet addressed whether there is an exception to the general fiduciary duty of confidentiality when a client makes serious criminal threats such as those in the present case. As previously noted, § 66(1) of the Restatement (Third) of the Law Governing Lawyers recognizes such an exception. I do not express a view about whether this jurisdiction should adopt the precise approach reflected in the Restatement. I would readily conclude, however, that on the foregoing facts defense counsel did not breach a fiduciary duty of confidentiality by disclosing Mr. 46 Moore’s repeated threats to shoot the prosecutor. To the contrary, in my view defense counsel acted entirely properly. See generally, e.g., Restatement (Third) of the Law Governing Lawyers § 66 cmt. b (“The exception recognized by this Section is based on the overriding value of life and physical integrity. . . . [I]t seems highly unlikely that a court would impose . . . liability on a lawyer” acting reasonably under Section 66(1) “to remove the threat to life or body”). Given that defense counsel had no fiduciary duty to keep Mr. Moore’s repeated threats confidential, Mr. Moore could not reasonably have expected defense counsel to keep the threats confidential. As the trial court noted, that is particularly obvious with respect to the second set of threats, which Mr. Moore uttered even after defense counsel had informed Mr. Moore that defense counsel would disclose Mr. Moore’s threats to the court if Mr. Moore repeated them. For the foregoing additional reasons, I conclude that Mr. Moore’s threats fall outside the scope of the attorney-client privilege. I briefly note one further topic. In addition to their fiduciary duty of confidentiality, attorneys also have an ethical duty of confidentiality. D.C. R. Prof. Conduct 1.6. That ethical duty is also not absolute, however, and D.C. R. Prof. Conduct 1.6(c)(1) permits disclosure of client confidences and secrets “to the extent reasonably necessary . . . to prevent a criminal act that the lawyer reasonably believes is likely to result in death or substantial bodily 47 harm absent disclosure.” We have indicated that this rule of professional responsibility “is not intended to govern or affect judicial application of the attorney-client privilege.” D.C. R. Prof. Conduct 1.6 cmt. [6]. It therefore is not strictly necessary in this case to decide whether defense counsel’s disclosure was permissible under R. 1.6(c)(1). Essentially for the reasons already stated, however, defense counsel’s disclosure in this case was in my view also clearly permissible under the Rules of Professional Conduct. DEAHL, Associate Judge, concurring: I agree that the attorney-client privilege does not protect the criminal threats that Brian Moore made to his attorney. I write separately first to explain why a straightforward application of the crime/fraud exception compels that result, and second to add some important caveats to the court’s alternative approach to the question at hand. A. The Crime/Fraud Exception Resolves This Case I’ll start with why the crime/fraud exception to the attorney-client privilege makes this an easy case. It is black letter law that communications between a client and their attorney are not privileged “if the attorney-client communication itself materially advances a crime or fraud.” In re Pub. Def. Serv. (In re PDS), 831 A.2d 890, 902 (D.C. 2003). A criminal threat uttered to one’s attorney does that. It not only advances, but completes, a criminal offense. Criminal threats thus make for an 48 easy application of the crime/fraud exception—Moore’s threats were not privileged because they advanced and consummated a criminal offense. I would end my analysis there. Moore and my dissenting colleagues offer two principal retorts to that straightforward analysis. Neither is persuasive. First, they invoke a principle that does not help them: that statements intended to advance a crime do not fall within the crime/fraud exception when “the attorney talks the client out of committing the crime.” In re PDS, 831 A.2d at 895. I agree with the principle, but Moore’s and the dissent’s invocation of it exhibits a fundamental misunderstanding of what a criminal threat is. A criminal threat is not an inchoate offense, or an incipient assault in the making that one can be talked out of. It is unto itself a completed crime upon the threat’s utterance to another party. United States v. Baish, 460 A.2d 38, 41 n.2 (D.C. 1983) (explaining that threats are no longer “inchoate” once “someone, other than the speaker, hears the threat”). So when a client communicates a true criminal threat to their attorney, they have not stated a mere intention to commit a crime, and they have not merely commenced a potential crime that their attorney can thwart. They have instead committed a complete offense. The attorney might try to talk their client out of committing 49 additional offenses by dissuading them from following through on the threat or repeating it. But the ship of talking them out of the criminal threat itself has sailed. Of course many seemingly threatening statements are not true criminal threats because a reasonable listener would not expect the speaker to follow through with them. And when in doubt, an attorney could always seek clarification of a client’s seemingly threatening statement and conclude that it was not a true threat where the client offers credible assurances that they would not follow through on it. So in that respect, we might say somewhat imprecisely that somebody can be talked out of an already-uttered threat, by which we mean they can assure the listener that it was not a true threat at its making. But that qualification does Moore no good here either— he declined repeated opportunities to assure his attorney, John Harvey, that his were not true threats, and instead repeated his death threats even after Harvey explained that he took them seriously and would disclose them if repeated. He simply was not talked out of his criminal offense no matter how you look at it, so Moore can take no shelter in espousing the virtues of giving attorneys the chance to talk their clients out of crimes. Harvey had that chance, and Moore was not talked out of anything. Second, Moore and my dissenting colleagues argue that the crime/fraud exception applies only where the attorney affirmatively assists their client in a criminal endeavor, and an attorney acting as a mere passive recipient of a threat does 50 not check that box. But this condition—let’s call it an assistance requirement—is largely of the dissent’s own invention rather than grounded in any authority. While the crime/fraud exception undoubtedly and commonly applies when the attorney affirmatively assists the client in a criminal endeavor, that is not a requirement for its application. All that is required for the crime/fraud exception to apply is for the communication, whether it comes from the client’s lips or the attorney’s, to have actually advanced a crime. 1 McCormick on Evidence § 95 (8th ed. 2022) (crime/fraud exception applies where “the communication was itself in furtherance of [a] crime or fraud”); Edward J. Imwinkelried, The New Wigmore: Evidentiary Privileges § 6.13.2(d)(1) (2d ed. 2009) (“The exception applies . . . when the communication itself was made in furtherance of the crime or fraud.”); see also In re Grand Jury Proceedings, 183 F.3d 71, 77 (1st Cir. 1999) (“[W]hen communications are intended to directly advance a particular criminal or fraudulent endeavor . . . their privileged status [will] be forfeited by operation of th[e crime/fraud] exception.”); West Virginia ex rel. Allstate Ins. Co. v. Madden, 601 S.E.2d 25, 37 (W. Va. 2004) (“The dispositive question is whether the attorney- client communications are part of the client’s effort to commit a crime or perpetrate a fraud.”). The dissent brushes by that black letter law because it is from other jurisdictions, suggesting that this court forged its own distinct path in In re PDS. 51 Post at 91-95. I see no reason why we should narrow our focus to our own precedents, which are no more binding on this court while sitting en banc than those from other jurisdictions—they are all exactly as persuasive as their reasoning. But even if I take the dissent up on its invitation to narrow the playing field down to its single preferred case, and ignore the overwhelming persuasive authority from elsewhere, the dissent simply misreads In re PDS. What In re PDS actually says is that the crime/fraud exception applies “if the attorney-client communication itself materially advances a crime or fraud,” In re PDS, 831 A.2d at 902 (emphasis added). It repeats that same point at least two more times, reiterating that the crime/fraud exception applies if a crime or fraud is “materially advanced by the communication itself,” id. at 907, and that “the attorney-client privilege is appropriately forfeited ab initio [when] the communication itself advances a crime or fraud.” Id. at 909. I thus do not see how In re PDS could have been any clearer that where the client’s communication itself advances a crime, as Moore’s threats did here, there is no additional requirement that the attorney further assist in the crime’s commission. The dissent counters that I have read each of those repeated statements “out of context,” because In re PDS provides an example of when client communications advance a crime, to wit, “as where the attorney agrees to help the client carry out his illegal scheme instead of rejecting it outright.” Id.; post at 92-94 (italicizing this language). But the phrase “as where” clearly introduces an example rather than 52 announcing a limitation on the previously articulated principle. The dissent repeatedly conflates the two things: it excerpts sentences from In re PDS that say the crime/fraud exception applies where the attorney assists his client in committing a crime, which is undoubtedly true, and reads them as if they are limitations on the exception, when they are not and cannot fairly be read that way. There are obviously times where client utterances themselves do nothing to advance a crime, as illustrated by the facts of In re PDS, where the entire scheme depends on the lawyer’s participation so that the lawyer’s refusal to participate “stops the client’s scheme dead in its tracks.” Id. at 895. In those instances, the attorney’s refusal to assist their client will preclude application of the crime/fraud exception because the client’s solicitation alone did not materially advance the crime—the attorney killed the crime in its cradle. But there are other times where a client’s utterances to a lawyer themselves advance or complete a crime, such as when they issue a criminal threat, and in those cases (like the present one) the crime/fraud exception applies. To be sure, it is a worse perversion of the privilege where the attorney actually helps their client commit a crime than when they are merely used as an unwitting sounding board for advancing it. But it is an abuse of the privilege all the same and the crime/fraud exception applies in either case, with or without the lawyer’s willing 53 assistance. 1 To hold otherwise would couple a capacious understanding of the attorney-client privilege with an incongruously and illogically narrow understanding of its limiting exception. Let me elaborate. The attorney-client privilege in its most skeletal form applies only where a client is seeking legal advice. A threat, of course, is not a request for legal advice; Moore even acknowledges in his brief that “[t]hreats . . . are not requests for advice.” So various courts have concluded—with some superficial appeal—that the privilege does not apply to threats without even addressing the crime/fraud exception. See, e.g., United States v. Ivers, 967 F.3d 709, 716-17 (8th Cir. 2020) (defendant’s threats to kill judge “were not for the purpose of obtaining legal advice” and so “are not 1 One might interject that an attorney would help their client commit the offense of a criminal attempt by even hearing out and rebuffing a prospective criminal scheme, which would put my view at odds with In re PDS. But that’s wrong: raising a potential criminal scheme with one’s lawyer does not amount to a criminal attempt. Mere contemplation or open discussion of a potential offense is not a “substantial step” that comes “dangerously close” to completing it. See Frye v. United States, 926 A.2d 1085, 1096 (D.C. 2005) (“To constitute an attempt, the act ‘must come dangerously close to completing the crime.’” (quoting Criminal Jury Instructions for the District of Columbia, No 4.04 (4th ed. 2002))); see also Sandoval v. Sessions, 866 F.3d 986, 989 (9th Cir. 2017) (“[M]ere solicitation . . . does not constitute ‘attempt[].” (citing United States v. Rivera-Sanchez, 247 F.3d 905, 908- 09 (9th Cir. 2001) (en banc))); United States v. Veliz, 800 F.3d 63, 72 n.9 (2d Cir. 2015) (“[M]ere solicitation is not an attempt.” (citing Wayne R. LaFave, Substantive Criminal Law § 11.1(f) (4th ed. 2003))). If, however, the client goes beyond merely communicating his scheme, and later takes a substantial step toward completing it with or without the lawyer’s assistance so that it does rise to the level of an attempt, then the crime/fraud exception indeed applies. 831 A.2d at 907 (rejecting view “that the crime-fraud exception applies only when a crime or fraud has been completed”). 54 protected by the attorney-client privilege”); United States v. Alexander, 287 F.3d 811, 816-17 (9th Cir. 2002) (defendant’s statements threatening violence “were clearly not communications in order to obtain legal advice” and were thus not privileged). Those courts have held that the privilege applies only when the “specific communication[]” at issue was strictly “necessary” to procuring discrete legal advice. Ivers, 967 F.3d at 717 (“Ivers’s statements . . . were in no way necessary to further his civil lawsuit.”); Alexander, 287 F.3d at 816 (“[T]he privilege is limited to only those disclosures . . . necessary to obtain informed legal advice.”). I agree with Moore and my dissenting colleagues that those cases reflect too impoverished a view of the attorney-client privilege because many attorney-client exchanges do not directly seek or offer discrete legal advice. That is especially true in criminal defense, where a critical part of the attorney’s job is gaining and maintaining a client’s trust. An effective attorney needs to have a holistic understanding of their client, complete with an understanding of their circumstances, worries, goals, and frustrations. Their communications thus must veer well outside the strict bounds of discrete legal advice, so that the privilege—if it is to be effective—must cover things like rapport-building and personal communications that foster and undergird the relationship. In the words of one preeminent evidence scholar, Professor Edward J. Imwinkelried: “For an attorney and a client to form an effective working relationship—one in which the client discloses all the facts in his 55 or her possession and develops enough trust in the attorney to take the attorney’s advice seriously—the attorney and client need ‘space’ in their interactions.” Parsing Privilege: Does the Attorney-Client Privilege Attach to an Angry Client’s Criminal Threat Voiced During an Otherwise Privileged Client Consulation, 72 Case W. Res. L. Rev. 871, 897 (2022). It would roughly obliterate the privilege if it protected only those individual statements that specifically seek, or furnish, legal advice. This court should firmly reject the unpersuasive reasoning in cases like Ivers and Alexander. But if you embrace a robust conception of the privilege, as I do, you should not couple it with a flaccid view of its limiting principle, the crime/fraud exception. That would pair an expansive view of the privilege as applying to client communications that are not truly seeking legal advice, like Moore’s threats, while incongruously straightjacketing the exception to those cases where advice is actually supplied. See supra n.1. That asymmetry is at the core of the dissent’s analysis. In my view, if one recognizes that the privilege extends to communications that do not directly seek advice, then one cannot coherently limit the crime/fraud exception to those instances where the attorney directly furnishes advice in response. The question instead should simply be whether the client’s communications sought to and ultimately did advance a crime, as Moore’s threats undoubtedly did here. 56 The policy rationales underlying the crime/fraud exception apply with full force when the communication at issue is itself a crime. Permitting clients to commit crimes in their attorney’s presence with impunity is precisely the type of “abuse[] of the attorney-client relationship to further the commission of a crime” that the crime/fraud exception was designed to prevent. In re PDS, 831 A.2d at 908. As Prof. Imwinkelried again persuasively explains after carefully walking through those policy justifications: “If anything, the case for recognizing a categorical exception for client statements amounting to illegal threats is stronger than the case for the well-settled crime/fraud exception to the attorney-client privilege,” because in the case of threats the crime is actually completed rather than merely advanced. Parsing Privilege, supra, 72 Case W. Res. L. Rev. at 904. I agree with him about that, for reasons the court’s opinion ably articulates, and I would hold that all criminal threats fit within the crime/fraud exception to the attorney-client privilege. It should nonetheless go without saying that defense attorneys are not obliged to divulge every imprudent client remark that might conceivably be taken as a threat, and no competent defense counsel needs to be told that. Emotions run high in criminal trials and defendants will sometimes vent in ways that a casual observer might interpret as a bona fide threat, but any halfway decent defense attorney will recognize as empty bluster. See State v. Boatwright, 401 P.3d 657, 664 (Kan. Ct. App. 2017) (An apparently threatening comment, while “jarring in isolation, . . . is 57 not an uncommon occurrence in the course of an attorney-client relationship.”). A defense attorney cannot be a pearl-clutching milquetoast, and it is the rare superficially threatening comment that is a true criminal threat, i.e., that gives rise to a reasonable belief “that the threatened harm would take place.” In re S.W., 45 A.3d 151, 155 (D.C. 2012). Here, Harvey testified that he had heard clients make facially threatening statements about prosecutors on “maybe three or four” prior occasions in his decades-long career and never previously felt obliged to disclose them. But in this case, Harvey made a judgment that Moore’s death threats— repeated after Harvey’s warnings that he took them seriously and would disclose them if reiterated—crossed the line into true threat territory. A jury agreed with that assessment beyond a reasonable doubt, a division of this court found the evidence sufficient to support that conviction, and we have no occasion to revisit that conclusion as the en banc court. I’m sure many defense attorneys would have made a different judgment call and chalked Moore’s comments up to venting or bluster, but the question before us is not whether Harvey acted as a model defense attorney. It is whether the attorney- client privilege protects true criminal threats uttered in the course of a representation, so that we effectively must assume Moore was reasonably likely to follow through on his death threats. Because such true threats are completed crimes, the crime/fraud exception dictates that the privilege’s protections do not extend to them. 58 B. A Few Caveats to the Court’s Alternative Rationale The court reaches a superficially more modest conclusion that criminal threats “to cause death or substantial bodily harm” are not privileged. Ante at 3, 10, 11, 18. I agree, though I would lose the quoted qualification, so I join the court’s opinion in full; it does not foreclose my broader view that all criminal threats are unprivileged. 2 The court reaches its conclusion after declining to engage with the standard two-step inquiry for analyzing attorney-client privilege issues. That is, it does not answer whether Moore’s communications generally fit within the attorney-client privilege in the first instance, or whether they fit within the crime/fraud exception in the second. It concludes only that threats of death or substantial bodily harm are an abuse of the privilege severe enough to be excepted from the privilege’s coverage due to “overriding policy concerns.” Ante at 14. 2 To illustrate where I would go further than the majority, consider a criminal defendant who tells his counsel that he will punch her in the face if he is convicted at trial. If that is a true threat of a simple assault that the attorney reasonably believes, my view is that it is not privileged. The majority remains agnostic on that point, at least if we understand a punch to the face as generally falling short of inflicting substantial bodily harm. The dissent, for its part, is conspicuously silent about this scenario. Post at 104 n.30. That silence surely stems from the fact that its otherwise hardline view becomes impossible to justify or even stomach if the privilege protects such direct threats to a lawyer. I believe such threats fall within the crime/fraud exception regardless of the threatened target. 59 I again agree because those overriding policy concerns are already baked into the privilege’s preexisting contours—criminal threats fit within the established crime/fraud exception. While I share the dissent’s distaste for truly new exceptions to a centuries-old doctrine, there is nothing new about the exception the court recognizes today, which is already encompassed by the crime/fraud exception. My only real issue with the majority’s analysis is that a more direct path is available to us, more firmly grounded in principles and precedent, as charted above. The court’s alternative path compels me to offer three caveats to its analysis. First is that its mode of analysis would not be viable if we were sitting as a division in this case. We have previously treated the crime/fraud exception for communications “made to further ongoing or future crimes” as the only exception to the attorney-client privilege. See, e.g., In re PDS, 831 A.2d at 906-08. The majority remains agnostic as to whether Moore’s threats fit within that long- recognized exception by crafting a new standalone exception for sufficiently serious threats. While we are free to do that as the en banc court—we could recalibrate the privilege entirely while sitting en banc, cabined only by the Supreme Court’s precedents and the Constitution itself—a division of this court would not be at liberty to do so. I think it is important to caveat that future divisions of this court are not free to emulate the majority’s analysis and craft new exceptions to the attorney-client privilege whenever they believe broader principles counsel in favor of that result. 60 They are instead bound to answer whether the statements at issue fit within the privilege and, if so, whether they nonetheless fit within a recognized exception to the privilege: (1) the crime/fraud exception, or (2) the newly-crafted serious threats exception (the second category adds nothing to the first, in my view). Second is that a majority of the en banc court, between the dissenters and I, has directly rejected the analysis in cases like Ivers and Alexander as unpersuasive. So notwithstanding the agnosticism expressed by the majority opinion on the topic, the controlling view resulting from today’s opinions is a rejection of the Ivers and Alexander approach that each discrete communication must directly seek legal advice for it to be covered by the privilege. That is a woefully anemic view of the attorney-client privilege that a majority of this court’s judges rightly rejects, if only through a combination of separate writings. Third, and finally, I would reaffirm the entirety of this court’s analysis in In re PDS, which persuasively explains why the crime/fraud “exception does not apply where the attorney talks the client out of committing the crime or fraud he contemplates or stops the client’s scheme dead in its tracks.” Id. at 895. I again take my dissenting colleagues to agree with me about that, so that a majority of this court has reaffirmed the continuing vitality of that precedent, in case anything in the 61 court’s opinion might be read to cast doubt on that. 3 Where I part ways with my dissenting colleagues is that it matters to me that unlike the client in In re PDS, Moore was not talked out of, but consummated, his criminal offense when he issued his threats. For that reason the crime/fraud exception takes Moore’s threats outside of the attorney-client privilege’s protections even without the new exception for serious threats that the court crafts. With that said, I concur. EASTERLY, Associate Judge, with whom BECKWITH and HOWARD, Associate Judges, join, dissenting: Evidentiary privileges, like the attorney-client privilege, exist to keep evidence out of court. By design, they counter the truth-finding function of our legal system. The attorney-client privilege has for centuries prohibited the use of attorney-client communications as evidence, despite their presumably valuable content, both because it has long been understood that regularly allowing an attorney to become a witness against their client would chill this societally valuable relationship and because, normatively, we do not want lawyers 3 I also share the dissent’s procedural concerns that Harvey probably should not have been allowed to testify about these client communications without giving Moore notice and an opportunity to litigate the privilege question. But that issue has not been raised or briefed before us, so I express no firm opinion on the matter and remain open to revisiting this tentative view in an appropriate case. 62 in our adversarial legal system to serve both as advocates for and witnesses against their clients. The privilege applies when an individual is communicating confidentially with their attorney to seek legal counsel. To date, the only exception this court has recognized to the privilege has been for communications that misuse the attorney- client relationship by furthering the commission of a crime. In this case, Mr. Moore was confidentially seeking guidance from his lawyer when, in the midst of his contempt trial, he spoke to his lawyer alone in the courthouse hallway about his extreme frustration with the efforts by the prosecutor to put him on a GPS monitor and his fear of losing his job as a result. Accepting the record as it comes to this court, but see Part III.A (questioning whether counsel should have been permitted to testify in the grand jury before Mr. Moore was given the opportunity to invoke the privilege), Mr. Moore made conditional statements in the course of his venting that he would shoot the prosecutor if she didn’t stop “fucking with him” or if he lost his job. Simply by making himself available to hear what Mr. Moore had to say (the content of which counsel could not have anticipated), counsel did not help Mr. Moore commit a crime. Under the longstanding common law test employed by this court, the privilege protected the statements Mr. Moore made to his attorney, and the attorney should not have been permitted to become a witness against his client at Mr. Moore’s subsequent trial for criminal threats. 63 The majority does not dispute that the attorney-client privilege in its existing form protected Mr. Moore’s statements to his lawyer and does not seek to defend the trial court’s erroneous ruling to the contrary. Ante majority at 11, 18. (Judge Deahl’s concurrence expressly rejects the trial court’s view of the privilege. Ante Deahl, J., concurring at 53-55, 60.) Instead, the majority creates a never-before- seen exception to the centuries-old attorney-client privilege that carves out confidential statements, made by a client to their lawyer while seeking legal advice, that could be perceived as threats of death or serious bodily injury. This novel exception is ill-founded. It is based partly on a misperception that lawyers who have challenging conversations with their clients are “abused” when those conversations cause any of a range of emotions and partly on the asserted concern that attorneys will be used as conduits for statements that could be perceived as threats. But having emotional and sometimes disconcerting conversations with clients is a core part of the job of a lawyer in direct client service, and to hold otherwise not only projects weakness on those lawyers, but also undermines their ability to zealously advocate for their clients and weakens the foundation of the privilege itself. Likewise, the majority fails to demonstrate how the mere possibility that the subject of a threat may later learn of the communications, against the client’s wishes, indicates that the client abused the attorney-client relationship by making those statements. Indeed, the majority does not require that the communication be 64 disclosed by counsel, much less transmitted to the subject, before its new exception applies. The majority’s exception is without analogue. No other jurisdiction in the country has a categorical threats exception to the attorney-client privilege. Lastly, the majority’s threats exception portends a grave intrusion on the attorney-client relationship, if not its eventual demise. Sitting en banc, we must look beyond the facts of this case and focus on the rule that is being announced and its effect in future cases. The majority’s threats exception is ominously expansive. This exception requires nothing more of counsel than that they do their normal job of listening to their client; it does not require counsel to disclose their client’s statements, much less transmit the statements to their target. The majority’s characterization notwithstanding, the exception is not limited to the revelation of criminal threats; rather, it casts a far wider net, given the majority’s explication that a judge need only “preliminarily” determine that a criminal threat was uttered— which requires some far lesser showing than guilt beyond a reasonable doubt. Ante majority at 40. The exception has no real-time or even future public-safety objective. It does not require that anyone—the attorney who heard the statement, the party who invokes the exception, or the judge who rules on its application—actually believe 65 the client intended to carry out the threat, and it allows for the admission of otherwise privileged statements in all types of cases, not just criminal cases. Thus, as crafted, the majority’s new exception creates a pernicious tool for government and private lawyers to exploit confidential communications between opposing counsel and their clients. With such a rule, the chilling of attorney-client communications, in particular between attorneys and clients seeking direct legal services in the matters that consume the District’s courts and agencies, is inevitable. Any good lawyer will caution their clients to regulate their speech. In so doing, the lawyer will send the now-accurate message that their conversations are not truly protected and that clients should think twice about revealing their thoughts and feelings to counsel. I. The Nature of Privileges and the Attorney-Client Privilege in Particular The maxim that, “[i]n our judicial system, the public has a right to every man’s evidence,” ante majority at 12 (quoting Trump v. Vance, 591 U.S. 786, 791 (2020)), is just that—a general principle, not an accurate descriptor of reality. In fact, swaths of evidence are kept out of our courts, regularly, for myriad reasons. In criminal cases, for example, defendants cannot be forced to incriminate themselves, the Confrontation Clause generally forbids the admission of testimonial statements in court by an absent speaker, and the exclusionary rule bars the government from 66 presenting evidence that it acquired by unconstitutional means. 1 And in all cases— civil and criminal—evidentiary rules like rules against hearsay, character evidence, and evidence that is substantially more prejudicial than probative block evidence from being considered by the factfinder. Privileges also keep evidence from the factfinder in civil and criminal cases. But whereas evidentiary rules are usually aimed at excluding unreliable evidence and promoting the truth-seeking function of a legal proceeding, privileges “are not designed or intended to facilitate the fact-finding process or to safeguard its integrity.” 1 McCormick on Evidence § 72 (9th ed.). “[R]ather than facilitating the illumination of truth, they shut out the light.” Id.; see also Rebecca Wexler, Ignorance of the Rules of Omission: An Essay on Privilege Law, 76 Vand. L. Rev. 1609, 1616 (2023) (stating that it is “clear” privileges are “unapologetically anti- accuracy”). Privileges reflect a determination that certain values or ends extrinsic to the litigation process are more important than accessing all the evidence or getting the “right” result in a particular case. Edward J. Imwinkelried, The New Wigmore: Evidentiary Privileges § 1.3 (2d ed. 2009) (defining a privilege as a “[s]tatutory or common law . . . [p]rocedural rule . . . inspired primarily by extrinsic social policy 1 See, e.g., Malloy v. Hogan, 378 U.S. 1, 6 (1964) (discussing the protection against self-incrimination); Crawford v. Washington, 541 U.S. 36, 68 (2004) (discussing the confrontation clause); Wong Sun v. United States, 371 U.S. 471, 484- 85 (1963) (discussing the exclusionary rule). 67 (as opposed to institutional concerns about the quality of judicial evidence or the functioning of the adversary legal system[)]” (footnotes omitted)); Wexler, supra at 1616 (explaining that privileges “reflect a judgment that certain values—be they personal privacy, commercial innovation, or military espionage—matter more than exposing truth in the courts”); Alex Stein, The New Doctrinalism: Implications for Evidence Theory, 163 U. Pa. L. Rev. 2085, 2094 (2015) (noting that privileges “suppress evidence for purposes unrelated to factfinding”). At common law, courts developed a subset of privileges to protect relationships—for example, between an individual and their clergyperson or their spouse—that were “regarded as of sufficient social importance to justify some sacrifice of availability of evidence relevant to the administration of justice.” 1 McCormick on Evidence § 72 (9th ed.); see also Imwinkelried, supra § 1.1 (“[U]nlike most evidentiary rules, privileges protect interpersonal relationships outside the courtroom.” (quoting Federal Rules of Evidence, H.R. Rep. No. 93-650, 93d Cong., 1st Sess. 28 (1973))); 8 Wigmore, Evidence § 2285 (McNaughton rev. 1961) (acknowledging that privileges should be created if there is a “relation” which the community believes “ought to be sedulously fostered”). “The rationale traditionally advanced for these privileges is that public policy requires the encouragement of the communications without which these relationships cannot be effective.” 1 McCormick on Evidence § 72 (9th ed.). But these privileges were also 68 supported by normative rationales, rooted in values such as privacy and autonomy, independent of the privileges’ effects on human behavior. Id. The oldest of these relational privileges is the attorney-client privilege. 2 Like other relational privileges, the attorney-client privilege is supported by a range of justifications. Wigmore championed the idea that “[i]n order to promote freedom of consultation of legal advisors by clients, the apprehension of compelled disclosure by the legal advisors must be removed; hence the law must prohibit such disclosure except on the client’s consent.” Wigmore, supra § 2291. Subsequent commentators have given this rationale a “utilitarian” or “instrumental” label, see, e.g., 24 Wright & Miller, Fed. Prac. & Proc. Evid. § 5472 (1st ed.), and unpacked its logic as follows: (1) “the law is complex”; (2) “it is in the public interest that citizens should understand the law and this can best be accomplished by allowing them to obtain advice from persons learned in the law”; (3) for attorneys to give good advice, they “must be informed by the client of all of the facts, good and bad” and, equally important, the client “needs to feel that he can tell his lawyer everything because this engenders trust and confidence in the lawyer,” which is “good in itself” and “makes 2 Wigmore dated the attorney-client privilege back to “the reign of Elizabeth I” (which spanned from 1558 to 1603), explaining that the privilege “commended itself at the very outset as a natural exception to the then novel right of testimonial compulsion.” Wigmore, supra § 2290; see also 1 Paul R. Rice, Attorney-Client Privilege: State Law Maryland § 1:2 (July 2024 Update). 69 it more likely that the client will accept the advice offered or approve the litigation strategy followed by counsel”; (4) without the privilege, clients would limit communications with counsel; and (5) thus, the benefits of the privilege outweigh the costs. Id. (footnotes omitted). In addition to this utilitarian rationale, the privilege is undergirded by the normative understanding that the revelation of attorney-client confidences is simply “wrong,” Wright & Miller, supra § 5472, in the nature of a “betrayal,” id. (quoting Zephaniah Swift, The Law of Evidence 94 (1810)), and contrary to “the notion that human dignity is one of the core values of [our] adversary system,” id. 3 Early commentators justified the privilege by appealing to a sense of morality: “To creep into the confidence of even a guilty man, under the pretence of being his defender, for the purpose of bringing him to justice, is an act that never can be viewed in a moral or respectable light.” Wright & Miller, supra § 5472 (quoting 1 Livingston, Works 465 (1873)); see also id. (“‘every feeling of justice, honour, and humanity would be shocked’ by the attorney’s revelation of his client’s secrets.”). Modern commentators are more likely to invoke the language of autonomy, dignity, privacy, 3 Although Wigmore argued that the normative rationale predominated only at the outset, and that it was soon replaced by the utilitarian rationale, see Wigmore, supra § 2290, that characterization was more aspirational than descriptive. “The elimination of non-instrumental justifications was a goal of Wigmore and the other Progressive writers, but it was one that was never achieved, except in their own writings.” Wright & Miller, supra § 5472 (footnote omitted). 70 and constitutional rights. Id.; see also Imwinkelried, supra §§ 1.1, 5.3.3(3)(c); 1 McCormick on Evidence § 72 (9th ed.). In criminal cases, the attorney-client privilege is now recognized as “a necessary concomitant of [a defendant’s] Fifth Amendment right not to incriminate himself and his Sixth Amendment right to the assistance of counsel.” Wright & Miller, supra § 5472; see also Neku v. United States, 620 A.2d 259, 262 (D.C. 1993) (“In the criminal context the [attorney-client] privilege acquires Sixth Amendment protection.”); accord In re Pub. Def. Serv., 831 A.2d 890, 899-900 (D.C. 2003) (hereinafter In re PDS). And “[s]ome courts have suggested that the privilege may receive some constitutional sanction in civil cases under the client’s right of privacy.” Wright & Miller, supra § 5472. In short, the privilege exists both to promote open and trusting attorney-client communication and to protect the role of lawyer as a “fighter for” their client, a “tradition [that] would be outraged by routine examination of the lawyer as to the client’s confidential disclosures regarding professional business.” 1 McCormick on Evidence § 87 (9th ed.). 71 II. The Attorney-Client Privilege in the District To determine when the attorney-client privilege applies, this court, like the federal courts, still employs the test formulated by Wigmore. 4 Under this test, the privilege applies: (1) where legal advice of any kind is sought (2) from a professional legal advisor in his capacity as such, (3) the communications relating to that purpose, (4) made in confidence (5) by the client, (6) are at his instance permanently protected (7) from disclosure by himself or by the legal adviser, (8) except the protection be waived. Jones v. United States, 828 A.2d 169, 175 (D.C. 2003) (quoting Wigmore, supra § 2292); see also ante majority at 13. When the privilege applies under this test, it has been “traditionally deemed worthy of maximum legal protection.” In re PDS, 831 A.2d at 900 (quoting Haines v. Liggett Grp. Inc., 975 F.2d 81, 90 (3d Cir. 1992)); 8 Wright & Miller, Fed. Prac. & Proc. Civ. § 2017 (3d ed.) (explaining that communications within the scope of the attorney-client privilege are “zealously protected”); see also Swidler & Berlin v. United States, 524 U.S. 399, 410 (1998) 4 See, e.g., Cavallaro v. United States, 284 F.3d 236, 245 (1st Cir. 2002); United States v. Stern, 511 F.2d 1364, 1367-68 (2d Cir. 1975); United States v. El Paso Co., 682 F.2d 530, 538 n.9 (5th Cir. 1982); Radiant Burners, Inc. v. Am. Gas. Ass’n, 320 F.2d 314, 319 (7th Cir. 1963); Matter of Fischel, 557 F.2d 209, 211 (9th Cir. 1977); see also Fisher v. United States, 425 U.S. 391, 403 (1976). 72 (expressing concern about a “‘no harm in one more exception’ rationale [that] could contribute to the general erosion of the privilege”). Until now, this court has recognized only one exception to the application of the privilege when all its conditions are met. Under the narrow crime-fraud exception, communications that are in “furtherance of an ongoing or future crime or fraud” are not protected. In re PDS, 831 A.2d at 902; see also id. (explaining that the exception applies “if the client uses the attorney’s advice or services to pursue a crime or fraud, or if the attorney-client communication itself materially advances a crime or fraud, even if the client’s efforts are frustrated or halted short of consummation of the evil deed”). But this exception has no application to communications regarding past wrongdoings, id., or communications that are not “actually . . . in furtherance of an ongoing or future crime,” id. at 895; see also id. at 906 (“It does not suffice that the communications may be related to a crime . . . . [T]he court must determine that the communication was itself in furtherance of the crime or fraud . . . .” (internal citations and quotation 73 marks omitted)). 5 In short, for the crime-fraud exception to apply, the attorney-client communication has to help the client advance a criminal or fraudulent purpose. 6 The crime-fraud exception is based on the understanding that the privilege should not apply where the attorney-client relationship is being “misused” or “abused” to advance criminal ends. Id. at 909-10; see also id. at 908 (explaining that the purposes of the attorney-client privilege “do not include concealing abuses of the attorney-client relationship to further the commission of a crime or fraud”); Clark v. United States, 289 U.S. 1, 15 (1933) (explaining that “[t]here is a privilege 5 See also Wigmore, supra § 2298 (“It has been agreed from the beginning that the privilege cannot avail to protect the client in concerting with the attorney a crime or other evil enterprise. This is for the logically sufficient reason that no such enterprise falls within the just scope of the relation between legal advisor and client.” (emphasis added)); Imwinkelried, supra § 6.13.2(d)(1) (“During an attorney-client consultation, the client might mention his or her planned future crime without seeking the attorney’s assistance in executing the plan. If so, the privilege would still attach. The exception applies only when the communication itself was made in furtherance of the crime or fraud.” (footnotes and internal quotation marks omitted)). 6 In re Sealed Case, 754 F.2d 395 (D.C. Cir. 1985), is a quintessential example of the application of this exception. In that a case, a church involved in litigation “embarked on a massive and systemic program to destroy and alter subpoenaed evidence or evidence sought pursuant to civil discovery requests,” with the help of outside counsel who facilitated the presentation of perjured testimony and filed (and verified the authenticity of) false documents. Id. at 397, 402. Based on its determination that counsel “were instrumentalities in the ongoing cover-up whether they realized it or not,” and the attorneys’ “representation and advice . . . assisted [the church] in carrying out its illegal and fraudulent scheme,” the court held that the crime-fraud exception applied and that counsel could be compelled to testify before a grand jury. Id. at 402. 74 protecting communications between attorney and client. The privilege takes flight if the relation is abused. A client who consults an attorney for advice that will serve him in the commission of a fraud will have no help from the law.”). In fulfilling their obligation to serve as zealous advocates, lawyers may help their clients evade civil and criminal liability by wielding all the tools of our adversarial system. But they cannot help their clients commit new crimes. III. The Application of the Privilege in Its Current Form to Mr. Moore’s Communications with Mr. Harvey The communications at issue in this case satisfied the criteria for the attorney- client privilege in its current form, and the crime-fraud exception has no application. Thus, under the law as it stood before the issuance of this opinion, Mr. Harvey should not have been permitted to testify against Mr. Moore at Mr. Moore’s later criminal trial for threats. The majority does not contend otherwise. Ante majority at 11, 18. There would be no need for it to create its new threats exception to the privilege if it did. A. Procedural Preface About the Communications at Issue Ordinarily, parties litigate whether the attorney-client privilege protects sought-after communications before they are disclosed for use in court, cf. Mohawk 75 Indus., Inc. v. Carpenter, 558 U.S. 100, 110-12 (2009) (explaining that once an order denying the protection of the privilege has been issued, a party can challenge that order before those communications are elicited as evidence, which “provide[s] assurances to clients and counsel about the security of their confidential communications”). Thus, for example, in In re PDS, the parties litigated the application of the privilege in the context of a motion to quash a subpoena for counsel’s grand jury testimony, which was immediately appealed. 831 A.2d at 895- 97. This court explained that “the party seeking to overcome the claim of privilege[] has the burden of proof,” and to meet that burden that party “must present the court with evidence showing probable cause to believe the crime-fraud exception applies to the communications in question.” Id. at 902. Observing in In re PDS that the government had sought to carry that burden with “an unsworn narrative proffer,” this court held that “an evidentiary submission, such as a sworn affidavit by a competent affiant, should be the norm.” Id. In this case, Mr. Harvey testified before a grand jury, an arm of the court, United States v. Perkins, 433 F.2d 1182, 1187 n.38 (D.C. Cir. 1970), giving a detailed narrative of Mr. Moore’s two sets of threatening statements in April and June 2018, without anyone inquiring whether his testimony should be barred by the attorney-client privilege. Mr. Moore was not given the opportunity to litigate whether the privilege protected his communications with Mr. Harvey from 76 evidentiary use until he went to trial on threats charges, 7 at which point the government used Mr. Harvey’s grand jury testimony to support its argument that the attorney-client privilege did not apply to Mr. Moore’s communications with Mr. Harvey. This was wrong. The communications between Mr. Moore and Mr. Harvey were presumptively privileged. See In re PDS, 831 A.2d at 903 (“The privilege presumptively applies both to the verbal statements of Attorney and Client and . . . to any documents conveyed by Client to Attorney for the purpose of securing legal advice.”). Arguably the government or the court had some obligation either to ensure that Mr. Moore had some ability to invoke the privilege or to invoke it on his behalf. 8 See Restatement (Third) of the Law Governing Lawyers § 86(1)(c) (Am. L. Inst. 2000) (providing that “the tribunal has discretion to invoke the privilege” where no one else has done so). But in any event, Mr. Harvey had an obligation as 7 As the majority acknowledges, when finally given the opportunity to contest the admission of Mr. Harvey’s testimony, Mr. Moore asserted that Mr. Harvey had “flat[]out lied” about their communications and that he had never made threatening statements about the contempt prosecutor to Mr. Harvey. Ante majority at 7-8 (alteration in original). 8 Notably, the government took care to advise Mr. Harvey of his rights before he testified before the grand jury. 77 Mr. Moore’s former lawyer to claim the privilege on Mr. Moore’s behalf and allow litigation (or waiver) of this issue before Mr. Harvey testified before the grand jury. As recognized in the Restatement of the Law Governing Lawyers, “[a] lawyer . . . from whom a privileged communication is sought must invoke the privilege when doing so appears reasonably appropriate, unless the client . . . has waived the privilege or . . . has authorized the lawyer . . . to waive it.” Id. § 86(1)(b); see, e.g., In re PDS, 831 A.2d at 896 (agency employing counsel moved to quash subpoena calling for counsel’s testimony before a grand jury on grounds of attorney- client privilege). 9 This obligation persists even after the attorney-client relationship has ended. See Wigmore, supra § 2292 (explaining privileged communications are “permanently protected . . . from disclosure”); Restatement (Third) of the Law Governing Lawyers § 77 (Am. L. Inst. 2000) (“Unless waived . . . or subject to 9 See also In re Sealed Case, 141 F.3d 337, 377 (D.C. Cir. 1998) (explaining that a lawyer is “normally duty-bound to assert” “the privilege of [their] client”); In re Search Warrant Issued June 13, 2019, 942 F.3d 159, 173 (4th Cir. 2019) (explaining that, “in proceedings such as these [litigating a challenge to government inspection of arguably privileged documents], lawyers are obliged to protect the attorney-client privilege to the maximum possible extent on behalf of their clients,” and stating that a “lawyer has a duty to invoke [a] claim of privilege on [their] client’s behalf”); Schwimmer v. United States, 232 F.2d 855, 863 (8th Cir. 1956) (“[T]he attorney has the duty, upon any attempt to require him to testify or produce documents within the confidence, to make assertion of the privilege, not merely for the benefit of the client, but also as a matter of professional responsibility in preventing the policy of the law from being violated.”); accord Tillotson v. Boughner, 350 F.2d 663, 665 (7th Cir. 1965); Hunter v. Kenney, 422 P.2d 623, 624- 25 (N.M. 1967). 78 exception . . ., the attorney-client privilege may be invoked . . . at any time during or after the termination of the relationship between client or prospective client and lawyer.”); id. § 77 cmt. c (“A lawyer for a client who has died has a continuing obligation to assert the privilege.”). 10 Litigating the privilege issue without Mr. Harvey’s grand jury testimony and with only the contemporaneous transcripts of Mr. Moore’s contempt trial in 2018 would have put this case in a very different light. Without Mr. Harvey’s after-the- fact characterizations in the grand jury, the contemporaneous transcripts make it far from clear that Mr. Harvey believed Mr. Moore had uttered any statement that constituted a criminal threat or posed a danger—and instead paint a picture of an attorney and client who simply did not get along. 10 See also United States v. Edgar, 82 F.3d 499, 508 (1st Cir. 1996) (explaining in the context of former counsel being subpoenaed to testify before a grand jury: “Generally, an attorney has an obligation to assert the privilege on behalf of the client and not to disclose confidential information until there is a judicial determination that there is no privilege. . . . Even if there is an assertion that there is no privilege because the crime-fraud exception applies, the attorney is required to give notice to the client.”); Republic Gear Co. v. Borg-Warner Corp., 381 F.2d 551, 556 (2d Cir. 1967) (holding in a case involving the invocation of the privilege on behalf of a former client that, “[n]ot only may an attorney invoke the privilege [o]n his client’s behalf when the client is not a party to the proceeding in which disclosure is sought, . . . but he should do so, for he is ‘duty-bound to raise the claim in any proceeding in order to protect communications made in confidence.’” (quoting A.B. Dick Co. v. Marr, 95 F. Supp. 83, 101 (S.D.N.Y. 1950))); accord McCormick on Evidence § 92 nn.8, 14 (9th ed.). 79 After the first set of alleged threats, Mr. Harvey told the court that he might “have an ethical issue” and asked for time to speak to “Bar Counsel,” seemingly referring to the D.C. Bar’s Legal Ethics Counsel. 11 Thereafter, he sought to withdraw from the case under D.C. R. Prof. Conduct 1.16(b), which provides standard for terminating representation. Although the court pressed him for more details, he repeatedly represented that he could not give the court more information, even though under the ethical rules he could have made a disclosure if in fact he “reasonably believe[d]” Mr. Moore was going to shoot the prosecutor. D.C. R. Prof. Conduct 1.6(c) (authorizing an attorney “to reveal client confidences . . . to the extent reasonably necessary . . . to prevent a criminal act that the lawyer reasonably believes is likely to result in death or substantial bodily harm absent disclosure.”). 12 11 “Bar Counsel” is the old name for the Office of Disciplinary Counsel. Office of Disciplinary Counsel: Purpose and Mission, DC Bar, https://www.dcbar.org/attorney-discipline/office-of-disciplinary-counsel/purpose- and-mission; https://perma.cc/D43X-XGRN. The D.C. Bar’s Legal Ethics Counsel operates a telephone helpline to provide real-time ethics advice for attorneys. See Ethics Advice, DC Bar, https://www.dcbar.org/for-lawyers/legal-ethics/ethics- advice; https://perma.cc/28DC-J5QY. 12 Even if he had made a Rule 1.6(c) disclosure, that would not have been dispositive of the application of the privilege. Because the duty of confidentiality differs in scope and serves distinct purposes from the attorney-client privilege, it is subject to its own distinct set of exceptions. See D.C. R. Pro. Conduct 1.6(c)-(e); see also id. cmt. 6 (explaining that the ethical rules are “not intended to govern or affect judicial application of the attorney-client privilege”). Moreover, although the confidentiality of a communication at the time it is made is an essential criterion of the privilege, the protection of the privilege is not lost if the attorney later breaches 80 Similarly, after the second set of alleged threats, Mr. Harvey’s focus was on withdrawing from his representation of Mr. Moore. After Mr. Moore’s alleged outburst, Mr. Harvey declared to the court: “I can’t represent him. And if you force me to tell you, then I’ll just have to tell you this time,” seemingly invoking D.C. R. Prof. Conduct 1.6(e)(2)(A) (authorizing disclosure inter alia “when required by . . . court order”)—not Rule 1.6(c). This statement prompted the court to issue a disclosure order, whereupon Mr. Harvey told the court that “he’s threatening to shoot her,” referring to the prosecutor, immediately followed by his declaration, “I ain’t representing him. We are adversaries, but I respect [the prosecutor]. I’m not representing that man.” His subsequent representations to the court arguably reinforce a conclusion that he was largely motivated to disclose Mr. Moore’s statements and pursue withdrawal because of the breakdown in his relationship with Mr. Moore—not because of his concern that Mr. Moore had uttered “an objectively credible threat.” In response to the court’s efforts to continue with the contempt trial the client’s confidence and discloses the communication to a third party. Only the possessor of the privilege—the client—can waive its protection. See Haines, 975 F.2d at 90 (“[The [attorney-client] privilege belongs to the client, and only the client may waive it.”); In re von Bulow, 828 F.2d 94, 100 (2d Cir. 1987) (“Of course, the [attorney-client] privilege belongs solely to the client and may only be waived by him.”); In re EchoStar Commc’ns Corp., 448 F.3d 1294, 1301 (Fed. Cir. 2006) (“This [attorney-client] privilege is at the discretion of the client.”); Wigmore, supra § 2321; D.C. R. Pro. Conduct 1.6 cmt. 7 (“The attorney-client privilege is that of the client and not of the lawyer.”). But the fact that Mr. Harvey did not make a Rule 1.6(c) disclosure provides strong evidence as to his contemporaneous view of Mr. Moore’s statements. 81 with Mr. Harvey as Mr. Moore’s counsel, 13 Mr. Harvey told the court that his relationship with Mr. Moore “ha[d] become toxic. Our ability to communicate–my skin starts to boil and so does his. I do not believe that it is a healthy relationship.” He further stated that his discussions with Mr. Moore “have been ugly but they’ve only been discussions” and that he was “not accusing Mr. Moore of anything other than the fact that he and my interactions have gotten to the point where[] both of us don’t want to be in the same room with each other.” Although the dissent recognizes that Mr. Moore has not raised this procedural issue, we detail this contemporaneous record information to underscore the procedural advantage the government gained by not being forced to litigate the privilege issue before it put Mr. Harvey before the grand jury. 14 Our aim is (1) to make clear that this court is not blessing the practice of putting a lawyer in front of a secret grand jury to reveal presumptively confidential client communications without giving the client a meaningful opportunity to invoke the attorney-client privilege, and (2) to highlight counsel’s obligation to invoke the privilege on behalf 13 The court was skeptical both that Mr. Moore’s statements had irreparably damaged Mr. Harvey and Mr. Moore’s relationship and, as the majority acknowledges, ante majority at 6-7, that Mr. Moore truly meant to hurt the prosecutor, see infra Part IV.A. 14 For its part, the government has not argued that the privilege was waived once Mr. Harvey testified in front of the grand jury, presumably recognizing how unfair it would be to deny Mr. Moore any opportunity to litigate this issue. 82 of current and former clients, particularly when those clients are not in a position to do so themselves. B. Mr. Moore’s Communications Related to a Legal Purpose. Before the division of this court that initially heard Mr. Moore’s appeal, the only question regarding the application of the attorney-client privilege was whether Mr. Moore was seeking legal assistance when he made the statements to Mr. Harvey that were subsequently prosecuted as criminal threats. 15 This question is easily 15 Before the en banc court the government argues for the first time that a different criterion—requiring the communications be made in confidence—was not satisfied with respect Mr. Moore’s second set of statements to Mr. Harvey, because Mr. Harvey had warned Mr. Moore that he would withdraw if Mr. Moore made any more threats. No member of this en banc court embraces this view, which would still render the first set of statements privileged. See ante majority at 11, 18; see generally ante McLeese, J., concurring at 42-47; ante Deahl, J., concurring at 47- 61. Judge McLeese asserts in his concurrence that neither of Mr. Moore’s statements were made in confidence, but for a different reason than the one briefed by the government: Mr. Moore could not have reasonably believed his statements would remain confidential because Mr. Harvey “had no fiduciary duty to keep Mr. Moore’s repeated threats confidential.” Ante McLeese, J., concurring at 46. Judge McLeese’s concurrence does not explain why the boundaries of the fiduciary duty should dictate the contours of the privilege, particularly where the boundaries of the ethical rule do not “govern or affect judicial application of the attorney-client privilege,” D.C. R. Prof. Conduct 1.6 cmt. 6; see also ante McLeese, J., concurring at 47, and the fiduciary duty and the ethical rule are so intertwined in our caselaw, see Griva v. Davison, 637 A.2d 830, 846-47 (D.C. 1994) (explaining that “a violation of the Code of Professional Responsibility or the Rules of Professional Conduct can constitute a breach of the attorney’s common law fiduciary 83 answered. Mr. Moore’s statements clearly related to a legal purpose—a proposition that no member of this en banc court contests. See ante majority at 11, 18; ante Deahl, J., concurring at 53-55, 60; see generally ante McLeese, J., concurring at 42- 47. As this court has previously explained, the attorney-client “privilege applies if one of the significant purposes of a client in communicating with a lawyer is that of obtaining legal assistance.” Jones, 828 A.2d at 175 (quoting Restatement (Third) of the Law Governing Lawyers § 72 reporter’s note (Am. L. Inst. 2000)); see also duty to the client”); Herbin v. Hoeffel, 806 A.2d 186, 195 & n.14, 197 n.17 (D.C. 2002) (citing the Rules of Professional Conduct for the standard of the fiduciary duty of confidentiality); id. at 197 (quoting a lawyer discipline case for the proposition that “disclosure of client confidences is ‘contrary to the fundamental principle that the attorney owes a fiduciary duty to [her] client and must serve the client’s interest with the utmost loyalty and devotion’” (quoting In re Gonzalez, 773 A.2d 1026, 1031 (D.C. 2001))). Furthermore, Judge McLeese’s concurrence also does not explain why Mr. Moore would not have “reasonably believed” his midtrial, private communications with his lawyer were made in confidence. Jones v. United States, 828 A.2d 169, 176 (D.C. 2003) (explaining that, when determining whether a communication was made in confidence, our focus is on what the client “reasonably believed” at the time the communication was made); see also ante McLeese, J., concurring at 42-43; Rule of Professional Conduct 1.6 cmt 7 (a client generally “has a reasonable expectation that information relating to the client will not be voluntarily disclosed and that disclosure of such information may be judicially compelled only in accordance with recognized exceptions to the attorney-client privilege and work product doctrine”). Even assuming Mr. Moore should be deemed to have had constructive knowledge of the boundaries of the fiduciary duty of confidentiality, as Judge McLeese acknowledges, our court has never addressed “whether there is an exception to the general fiduciary duty of confidentiality when a client makes serious criminal threats.” Ante McLeese, J., concurring at 45. 84 Wigmore, supra § 2310 (“[T]he client cannot know what is necessary or material . . . . The test is, therefore, not whether the fact or the statement is actually necessary or material or relevant to the subject of the consultation, but whether the statement is made as a part of the purpose of the client to obtain advice on that subject.”). So construed, the “significant purpose” test sensibly operates in many contexts to exclude communications that are only “tangentially related” to a legal purpose, Att’y Gen. of U.S. v. Covington & Burling, 430 F. Supp. 1117, 1121 (D.D.C. 1977), particularly where the attorney and the client are mutually engaged in both legal and nonlegal activities, see, e.g., Jones, 828 A.2d at 175 (“In the case of someone seeking advice from a friend who is also a lawyer, the lawyer-friend must be giving advice as a lawyer and not as a friend in order for the privilege to attach.”). Accepting Mr. Harvey’s narrative, Mr. Moore’s statements were made with the significant purpose of seeking legal assistance. Mr. Harvey was representing Mr. Moore in a contempt proceeding after Mr. Moore was accused of violating an order prohibiting him from having contact with his then-wife. Both sets of statements related to Mr. Moore’s conditions of release; Mr. Moore was expressing his strong emotions about the prosecutor’s efforts to restrict his liberty, as well as his fear of losing his job. These statements, intemperate though they may have been, fall within the heartland of the attorney-client privilege. 85 Mr. Moore made the first set of statements about which Mr. Harvey testified just after the prosecutor asked the court to order Mr. Moore to wear an ankle monitor as a condition of release. When Mr. Harvey and Mr. Moore walked out into the courthouse hallway after the hearing, Mr. Moore expressed his frustrations about the prosecutor and her request, telling Mr. Harvey, “Fuck that bitch. I hate this bitch.” Although Mr. Harvey acknowledged at trial that Mr. Moore’s feelings of anger were “normal,” Mr. Harvey told Mr. Moore at the time that it was “just silly on his part to be angry.” Not reassured by this response, Mr. Moore went further, saying, “Man, fuck that bitch. Fuck that bitch. I’ll shoot that bitch. Fuck that bitch.” 16 When Mr. Harvey asked, “Man, what are you talking about?” Mr. Moore responded, “That’s right, Harvey, I’ll shoot that bitch.” And when Mr. Harvey told Mr. Moore, “You starting to make me think you serious,” Mr. Moore answered, “God damn right, Harvey. Fuck that bitch. I’ll shoot that bitch.” 17 But shortly thereafter, 16 Before the grand jury, Mr. Harvey testified that Mr. Moore’s statements were more clearly conditional: “She keep fucking with me, I’m going to shoot this bitch. I’ll fuck this bitch up.” Mr. Harvey also testified that, in response to his attempt to defend the prosecutor (who Mr. Harvey identified as a friend to whom he spoke every day), Mr. Moore asked, “What, are you trying to have sex with her or something?” According to Mr. Harvey’s grand jury testimony, Mr. Moore also told him 17 he owned guns (presumably in North Carolina). Ante majority at 4. 86 Mr. Moore told Mr. Harvey that he was “just bullshitting” and “didn’t mean it.” 18 Mr. Harvey said he would believe Mr. Moore if Mr. Moore ever “use[d] this kind of language” again. Mr. Harvey further admonished Mr. Moore, “If you decide you want to go shoot somebody, you need to keep that to yourself and don’t make me a part of it.” 19 Mr. Moore assured Mr. Harvey that he would “say nothing like that again” and repeated that he “was just bullshitting.” Between this incident and the next, Mr. Moore started training for a new job as a crane operator. This change made it even more difficult for Mr. Moore to travel from North Carolina for his court dates because he was required to attend a certain 18 Mr. Harvey had seemingly already reached this conclusion, given that, as noted above, see supra Part III.A, he asked the court to withdraw pursuant to Rule 1.16. Even after consulting with Legal Ethics Counsel to determine the circumstances under which he could disclose Mr. Moore’s statements to the court, see ante majority at 4, he did not make a Rule 1.6(c) disclosure and repeatedly asserted that he was not permitted to share what Mr. Moore had said absent a court order. The majority suggests that Mr. Harvey specifically invoked the provisions of Rule 1.16(b)(1) and (2), which allow counsel to “withdraw from representing a client if . . . [t]he client persists in a course of action involving the lawyer’s services that the lawyer reasonably believes is criminal or fraudulent . . . [or] the client has used the lawyer’s services to perpetrate a crime or fraud.” See ante majority at 4-5. Although Mr. Harvey told the court presiding over the contempt trial that he “ha[d] concerns” related to these rules, when he was asked at Mr. Moore’s threats trial if he had sought to rely on these two provisions, he responded “no.” Instead, he indicated that he was seeking to rely on “all five parts” of Rule 1.16(b) generally. 19 Before the grand jury only, Mr. Harvey testified that he told Mr. Moore that if Mr. Moore “ever sa[id] anything like that to [him] again,” Mr. Harvey would “tell the court that [he] said it.” 87 number of classes within a specified period of time. Mr. Harvey attempted to convince the judge to allow Mr. Moore to be absent from some of the hearings, but he was unsuccessful, and the court said that if Mr. Moore did not attend all of the hearings in person, it would issue a bench warrant for Mr. Moore’s arrest. At the next hearing, the prosecutor succeeded in convincing the court, over Mr. Harvey’s objection, to order Mr. Moore to wear an ankle monitor. But because the Friday afternoon hearing had run late, Mr. Moore was ordered to return on Monday for the monitor’s installation. Mr. Moore was extremely upset because he would have to stay in the District for multiple more days and miss one of his trainings on Monday. As soon as they exited the courtroom, Mr. Moore told Mr. Harvey, “Harvey, Harvey, if I lose my job, I’m going to bust a cap in this bitch, I’m going to bust a cap in this bitch.” When Mr. Harvey asked, “Man, what are you doing?” Mr. Moore responded “Man, fuck this bitch,” and repeated, “If I lose my job, I’m going to bust a cap in this bitch,” punctuating his statement with a shooting gesture. Mr. Harvey then said, “I told you what I was going to do if you ever said something like that to me again,” which prompted a “fuck her, fuck you” response from Mr. Moore. 20 There is no trial evidence that Mr. Harvey had told Mr. Moore what he 20 would do if Mr. Moore made these statements again. See supra note 19. As it 88 The heated manner in which Mr. Moore asked Mr. Harvey for help with his conditions of release does not defeat the application of the attorney-client privilege to his exchanges with counsel. Imposing formalistic requirements about how one must request advice for the privilege to apply would blink reality and be antithetical to the purposes of the privilege. Practicing attorneys represent their clients in all sorts of stressful situations where strong emotional reactions are to be expected. From a utilitarian perspective, we want people to voice their frustrations to their attorneys—including their frustrations about how their case is being handled. See supra Part I; see also Imwinkelried, supra § 6.11 (acknowledging that in any privileged relationship there must be some “‘space’ [for the relationship] to flourish” (citation omitted)); 1 Paul R. Rice, Attorney-Client Privilege in the United States § 5:21 (2023) (“[A]n open, dynamic, and unguarded atmosphere” is the “very goal” of the attorney-client privilege.); ante Deahl, J., concurring at 54 (agreeing that “[a]n effective attorney needs to have a holistic understanding of their client, complete with an understanding of their circumstances, worries, goals, and frustrations” and attorney-client “communications thus must veer well outside the strict bounds of discrete legal advice” and “cover things like rapport-building and personal happened, Mr. Harvey returned to the judge, told her he could not represent Mr. Moore, and said if she forced him to, he would disclose Mr. Moore’s statements to her. See supra Part III.A. 89 communications that foster and undergird the relationship”). 21 And, normatively, we want all laypeople—regardless of their innate abilities to navigate the legal system or their emotional resilience at any particular time—to be able to rely on expert legal advisors to level the playing field in our adversarial justice system. Because Mr. Moore’s statements met all the criteria for protection under the attorney-client privilege, Mr. Harvey should not have been permitted to testify about these statements unless an exception to the privilege applied. C. The Crime-Fraud Exception Does Not Apply. The only previously recognized exception to the attorney-client privilege in the District, the crime-fraud exception, does not apply to the statements Mr. Moore conveyed to Mr. Harvey. As noted above, attorney-client communications must be “in furtherance of an ongoing or future crime” to fall under this exception. In re PDS, 831 A.2d at 895; see also supra Part II. Here, the crime Mr. Moore was charged with committing was the crime of threats, which, as interpreted by this court, requires proof that the defendant (1) “uttered words to another person” (2) “with a 21 The dissent welcomes Judge Deahl’s affirmative embrace of this broad view of privileged communications and shares his understanding that a majority of the en banc court has rejected the “impoverished . . . view” of the privilege espoused by the Eighth Circuit in United States v. Ivers, 967 F.3d 709 (8th Cir. 2020), and the Ninth Circuit in United States v. Alexander, 287 F.3d 811 (9th Cir. 2002). Ante Deahl, J., concurring at 54; see also id. at 60. 90 result that ‘the ordinary hearer [would] reasonably . . . believe that the threatened harm would take place,’” and (3) “with the purpose to threaten or with knowledge that his words would be perceived as a threat.” 22 Carrell, 165 A.3d at 320, 324 (alterations in original) (footnote omitted) (quoting In re S.W., 45 A.3d 151, 155 (D.C. 2012)). But nothing that Mr. Harvey did or said furthered Mr. Moore’s commission of the crime of threats. He did not, for example, endorse and encourage repetition of Mr. Moore’s hotheaded statements, offer to convey them to the prosecutor, or give advice as to how Mr. Moore could more effectively intimidate the prosecutor. To the contrary, Mr. Harvey, in keeping with his duties as Mr. Moore’s counsel, attempted (albeit arguably ineffectually) to get Mr. Moore to stop his rants and calm down. Because Mr. Moore did not “obtain[] from his lawyer the assistance he s[ought] to commit a crime or fraud,” In re PDS, 831 A.2d at 909— he never even sought assistance from counsel for a criminal or fraudulent purpose; he was seeking assistance to address his conditions of release in his ongoing contempt case—his communications with Mr. Harvey do not fall within the crime- fraud exception to the privilege. Cf. United States v. Chase, 340 F.3d 978, 982 (9th Cir. 2003) (explaining that threats made by an individual to his therapist were privileged because “[o]nce [he] finished uttering the threats, the charged crime was 22 Whether proof of recklessness towards the possibility that one’s words would be perceived as a threat is sufficient remains an open question. Carrell v. United States, 165 A.3d 314, 324 (D.C. 2017). 91 completed, and the psychiatrist was in the same position she would have occupied had her patient described a bank robbery in which he had participated a week earlier”). Judge Deahl takes issue with this analysis and concludes in his concurrence that Mr. Moore’s statements to Mr. Harvey fall under the crime-fraud exception simply because Mr. Harvey heard them. But this view is based on a misunderstanding of In re PDS and an unpersuasive assertion that a “capacious understanding of the attorney-client privilege,” ante Deahl, J., concurring at 53, which the concurrence endorses, must be coupled with an equally capacious understanding of the crime-fraud exception. Judge Deahl’s concurrence disputes that In re PDS recognized what he labels “an assistance requirement” for the crime-fraud exception to apply and asserts that such a requirement is “largely of the dissent’s own invention.” Id. at 50. It is not. In re PDS expressly held that “the crime-fraud exception is applicable where a client ‘consults a lawyer for the purpose, later accomplished, of obtaining assistance to engage in a crime or fraud,’” meaning that “the privilege is lost once the client obtains from his lawyer the assistance he seeks to commit a crime or fraud.” 831 A.2d at 909 (emphasis added) (quoting Restatement (Third) of the Law Governing Lawyers § 82(a)); see also id. (explaining that “the exception should operate, for 92 example, where the attorney is duped into unwittingly facilitating the client’s unlawful scheme”); 23 id. at 910 (explaining that “[t]he government does not have to show that the intended crime or fraud was accomplished, only that the lawyer’s advice or other services were misused” (emphasis added)); id. (holding that the crime-fraud exception did not apply because the Attorney “renounced” the client’s plan in their communications and subsequently did not do “anything to advance Client’s fraudulent plan after they conferred”). 24 To support its claim that the crime-fraud exception recognized by In re PDS encompasses a client’s otherwise privileged, threatening statements to counsel simply because counsel heard them, the concurrence relies on language in In re PDS that communications between a client and their attorney are not privileged “if the 23 The concurrence’s criticism of the dissent for “limit[ing] the crime/fraud exception to those instances where the attorney directly furnishes advice in response” to a client’s communication is misplaced. Ante Deahl, J., concurring at 55. We understand In re PDS to hold that the crime-fraud exception applies once the attorney has assisted the crime or fraud in some way—not necessarily only by providing advice. In re PDS, 831 A.2d at 910 (explaining that “[t]he government does not have to show that the intended crime or fraud was accomplished, only that the lawyer’s advice or other services were misused” (emphasis added)). 24 This understanding of the crime-fraud exception aligns with its black letter conception, as evidenced by In re PDS’s reliance on the Restatement. See 831 A.2d at 906, 909; see also Imwinkelried, supra § 6.13.2(d)(1) (“During an attorney-client consultation, the client might mention his or her planned future crime without seeking the attorney’s assistance in executing the plan. If so, the privilege would still attach.”). 93 attorney-client communication itself materially advances a crime or fraud.” Ante Deahl, J., concurring at 47 (quoting In re PDS, 831 A.2d at 902). It then reasons that because a criminal threat, as soon as it is uttered to a client’s attorney, “not only advances, but completes, a criminal offense,” such a communication falls within the crime-fraud exception. 25 Id. But the concurrence reads the phrase “the communication itself materially advances a crime or fraud” out of context. In the quoted passage, this court explained that it was “reject[ing] the holding of some courts to the effect that the crime or fraud that is the subject of the [attorney- client] communication must thereafter be completed for the [crime-fraud] exception to apply”; instead, the court held that, “[b]roadly speaking, it is enough if the client uses the attorney’s advice or services to pursue a crime or fraud, or if the attorney- client communication itself materially advances a crime or fraud, even if the client’s efforts are frustrated or halted short of consummation of the evil deed.” In re PDS, 831 A.2d at 902. The court then explained that, by contrast, “an ill-motivated client communication that ‘goes nowhere’—as where the client consults an attorney with an evil purpose but the attorney quashes the venture before anything further is done to promote it—is not sufficiently in furtherance of a crime or fraud to fall within the 25 The majority declines to endorse the concurrence’s analysis, ante majority at 11, 19 n.4, although it asserts (mistakenly, see infra Part IV.A), that its new threats exception is conceptually similar to the crime-fraud exception, ante majority at 19. 94 crime-fraud exception.” Id. In other words, in this passage of the opinion, the court was focused on the attorney-client interaction and was explaining that, although successful completion of the criminal or fraudulent scheme was not required, the client at least had to gain something from his communication with counsel—some information, guidance, or other service—to advance the client’s illegal (if ultimately unsuccessful) aims. The court returned to this understanding of “materially advances” later in the opinion when it explained why the crime-fraud exception did not apply in that case. The court observed that “the government [had] proffered nothing to suggest that the alleged fraudulent scheme was either pursued after Client’s communications with Attorney or materially advanced by the communications themselves. There is no suggestion whatsoever that Attorney did anything, even unwittingly, that was conducive to the fulfillment of the alleged scheme.” Id. at 907 (emphasis added). As for what the court meant by a “communication [that] itself advance[d] a crime or fraud,” the court explained it was referring to “certain special cases . . . as where the attorney agrees to help the client carry out his illegal scheme instead of rejecting it outright.” Id. at 909 (emphasis added). The court gave no indication that the one- way communication of a client’s thoughts or feelings to their counsel, alone, would trigger the crime-fraud exception. 95 Without the support of In re PDS, the concurrence is left with its argument that “coupl[ing] a capacious understanding of the attorney-client privilege” with a narrow understanding of the crime-fraud exception would be “incongruous[]” and “illogical[].” Ante Deahl, J., concurring at 53. The dissent and the concurrence are aligned in our “embrace [of] a robust conception of the [attorney-client] privilege.” Id. at 55. But the argument that this conception of the privilege must be coupled with an equally broad conception of the crime-fraud exception fails to persuade. To date, we have recognized that when the privilege applies under the Wigmorian test, it is “worthy of maximum legal protection.” In re PDS, 831 A.2d at 900 (quoting Haines, 975 F.2d at 90). And for good reason: a privilege that is broad but porous cannot effectively serve the critical role of allowing full and frank discussion between attorney and client, nor does it advance the normative interests the privilege is meant to effectuate. To say that any lawyer who provides the audience for a threat materially advances that crime simply because they helped fulfill the essential element of the threat being communicated, see ante majority at 17 (citing United States v. Baish, 460 A.2d 38, 42 (D.C. 1983)), is inconsistent with those utilitarian and normative goals. Indeed, it would seem to suggest that the lawyer did something wrong simply by doing their job of listening to their client. That cannot be right. The act of listening to one’s client is the essential precursor to the delivery of the effective assistance of counsel. It is in large part because we so 96 value this act that we have the privilege. See supra Part I. We want clients to feel free to unburden themselves to their lawyers so that counsel can serve as the buffer, the filter, and the course-corrector. Likewise, we want lawyers to act as a sounding board for whatever a client has to say in pursuit of legal advice (which a lawyer will never be able to fully anticipate). Recognizing that “discouraging clients from illegal conduct is a regular occurrence in an attorney’s practice,” we want lawyers to take in their clients’ communications about their potential or actual illegal activity, contemplated or completed, and then to “urge the client, as forcefully and emphatically as necessary, to abandon illegal conduct or plans,” and plot a lawful course forward. In re PDS, 831 A.2d at 901; see also id. (explaining that a lawyer’s advice “is even more vital when the client misguidedly contemplates or proposes actions that the client knows to be illegal”); see also Nix v. Whiteside, 475 U.S. 157, 169 (1986) (noting that a lawyer’s “first duty . . . is to attempt to dissuade the client from [an] unlawful course of conduct”); Wigmore, supra § 2298 (explaining that “the privilege, in its very fundamentals, presupposes . . . the furnishing of legal advice to the culpable client”). Thus, we want even statements that could be perceived as threats to fall within the attorney-client privilege. Notwithstanding that the proverbial ship has sailed and the lawyer cannot counsel their client not to engage in a completed threat crime, the lawyer will still have the opportunity both to persuade the client not to act on their 97 threatening words and to think through their legal challenges more productively. (But under the concurrence’s conception of the crime-fraud exception, opposing counsel could compel disclosure of a client’s statements even when counsel was of the view that hearing the client out had given counsel the opportunity to build trust and redirect the client to conform his behavior with the law.) “To withhold the privilege from such communications would be a mistake, for it effectively ‘would penalize a client for doing what the privilege is designed to encourage—consulting a lawyer for the purpose of achieving law compliance.’” In re PDS, 831 A.2d at 909 (quoting Restatement (Third) of the Law Governing Lawyers § 82 cmt. c (Am. L. Inst. 2000)). Or to put it more bluntly, the last thing we want is for lawyers to tell their clients, as Mr. Harvey did in Mr. Moore’s case, “If you decide you want to go shoot somebody, you need to keep that to yourself and don’t make me a part of it.” True, Mr. Moore did not take that advice; despite this admonition, he vented to his lawyer again. But this data set of one does not mean that as a general matter attorney-client communications would not be broadly chilled by lawyers regularly telling their clients to keep their thoughts to themselves. 26 26 The concurrence states that Mr. Moore “can take no shelter in espousing the virtues of giving attorneys the chance to talk their clients out of crimes” because, 98 In sum, the concurrence’s arguments notwithstanding, Mr. Moore’s statements to Mr. Harvey do not fall under the crime-fraud exception as recognized by In re PDS. And to modify that exception to encompass Mr. Moore’s statements would be inconsistent with the broader principle espoused in In re PDS that the attorney-client privilege “should be construed so as to effectuate and be ‘consistent with [its] purposes.’” Id. at 908 (quoting Swidler & Berlin, 524 U.S. at 409-10). It does not align with the purposes of the attorney-client privilege to except from its protection any communication made in the course of seeking legal advice that completes a crime simply because the attorney heard it. Rather, it fully serves the purposes of the attorney-client privilege to apply it to communications that are themselves crimes, so long as the attorney does not do or say anything to advance the client’s criminal aims. IV. The Majority’s Novel Exception for Threats of Substantial Bodily Harm To recap, the majority does not defend the trial court’s determination that Mr. Moore was not seeking legal advice when he uttered the statements charged as even after Mr. Harvey tried to get him to calm down, Mr. Moore continued to make threatening statements. Ante Deahl, J., concurring at 49. But the concurrence’s focus on the crime of threats is too narrow. We make this point about the critical role of lawyers in promoting compliance with the law not only because we want lawyers to talk their clients out of making (or repeating) threats, but also, even more critically, because we want lawyers, even after a true threat is stated, to be in a position to convince clients not to cause actual death or serious bodily injury. 99 threats, and the concurrence rejects this ruling outright. Instead, the majority affirms on other grounds: it creates a never-before-seen exception to the centuries-old attorney-client privilege, carving out statements that, although made within the context of communications seeking legal counsel, could be perceived to threaten death or serious bodily injury. The latter term, which is not defined in our caselaw, seemingly derives from Rule 1.6(c) (allowing disclosure of such statements), but notably, disclosure by counsel under Rule 1.6(c) or for any other reason is not a prerequisite for the application of this exception. The majority’s novel exception is ill-founded, has no analogue in any other jurisdiction, and will weaken both attorney- client relationships and the attorney-client privilege. A. The Majority’s “Abuse” Rationale As discussed above, the only exception to the attorney-client privilege that this court has recognized—the crime-fraud exception—is based on the rationale that using one’s counsel to advance criminal or fraudulent schemes is an “abuse[] of the attorney-client relationship,” which should not be concealed. In re PDS, 831 A.2d at 908; see also supra Part II. The majority seeks to rely on this “abuse of the relationship” rationale to justify its new exception, asserting that threats uttered to counsel are abusive because they can both induce unpleasant emotions in counsel and, when passed on by counsel to the target of the threat, induce fear in that 100 individual as well. Ante majority at 18-23. But the majority’s “abuse of the relationship” rationale based on purported emotional harm to counsel is so broad as to render the concept meaningless, and to the extent the majority relies on indirect harm to third parties through counsel, it misses the mark because it fails to require any disclosure by counsel, much less disclosure to the target, as a prerequisite for its new exception. The majority envisions that, upon hearing a threat to a third party, the lawyer may experience any of a spectrum of adverse reactions, ranging from “discomfort” and “intimidation” to “fear,” but also including “outrage” or seemingly any “other strong emotion.” Id. at 20, 23. This concept of “abuse” of the attorney-client relationship is nonsensical and sows the seeds of the attorney-client privilege’s destruction. As anyone who does direct client representation knows, individuals generally do not retain or have lawyers appointed to them because their lives are sunshine and roses. They enter attorney-client relationships because something has gone wrong, often in a life-altering way: they have been charged with a crime; they are being evicted; they have lost their job or their benefits; they are getting divorced; their children are being taken away from them. These clients are often under enormous strain and may disclose all sorts of information to their lawyers that is discomfiting, 101 intimidating, or even frightening. Their wells of resilience, which may never have been terribly full to begin with, may have run dry, and they may understandably have frustrations and concerns that are not always expressed with complete calm or anything close to it. They may even have engaged in illegal activity in the past, which they must discuss in detail with their attorney—and those details may well be upsetting for the attorney to hear. 27 But good lawyers know that it is their job to meet their clients where they are; to allow their clients to reveal their thoughts and emotions while setting aside their own emotional reactions to that information; and to thereby connect with their clients and forge a trusting relationship. See ante Deahl, J., concurring at 54-55. In other words, encountering unpleasant information or strong emotions and managing one’s own reactions thereto is simply part of the job of engaging in direct client representation. Such conversations are not only to be tolerated but are essential to an effective attorney-client relationship. 28 Thus, to say that such 27 In the criminal context in particular, clients charged with the most serious crimes like rape and murder may talk to their lawyers about how and why they committed these crimes. But in so doing they do not abuse the attorney-client relationship even if the details they share are distressing. 28 For this reason, the majority entirely misses the mark when it observes that clients know that they cannot utter statements that might be perceived as threats “[o]utside of the context of [their] relationship with [their] attorney,” so they should know that they cannot make such statements to their attorney. Ante majority at 19- 20, see also id. at 34-35. Clients have conversations with their lawyers that they do 102 conversations are an abuse of the attorney-client relationship reveals a complete misunderstanding of direct client representation and the general caliber of the attorneys who perform this work. As a result, it poses an existential threat to the attorney-client privilege in our jurisdiction. If future en banc courts employ this understanding of abuse, 29 the privilege will be swallowed by its exceptions. The facts of this case reinforce the destructive breadth of the majority’s “abuse of the relationship” rationale. There is no question in this case that Mr. Harvey did not like what Mr. Moore said about the prosecutor in their conversations, but there is no evidence on this record that Mr. Moore’s statements ever caused Mr. Harvey to be fearful of or intimidated by Mr. Moore. And, as previously noted, Mr. Harvey never made a Rule 1.6(c) disclosure; rather, he repeatedly stated that he could not, even though Bar ethics counsel had told him “what [his] options were” under the not have with anyone else. They reveal unsavory, unfavorable, and perhaps even inculpatory information that they do not reveal to anyone else. The normal boundaries of “day-to-day” conversation do not dictate the boundaries of the attorney-client relationship. Likewise, the majority is wrong to say that the dissent “tells attorneys to toughen up.” Id. at 20 n.6. The dissent does not need to send that message. We know that the attorneys who represent clients in our court system are already up to the task. Instead, we challenge the majority to see attorneys who perform direct client representation as professionals—not, in the words of Judge Deahl, “pearl- clutching milquetoast[s].” Ante Deahl, J., concurring at 57. 29 We agree with Judge Deahl’s concurrence that only an en banc court may create a new exception to the attorney-client privilege. See ante Deahl, J., concurring at 59-60. 103 ethical rules, i.e., that he could make a disclosure if he reasonably believed that Mr. Moore posed a threat of death or serious bodily harm. See ante majority at 4. Instead, the record indicates that Mr. Harvey’s overriding emotion was anger, in particular about how Mr. Moore was talking about the prosecutor, a person he considered not just a “colleague” but a “friend.” Mr. Harvey told the trial judge that Mr. Moore made his “skin . . . boil”; that Mr. Moore “likes to cross certain lines that men don’t cross”; and that he had responded to Mr. Moore in a “very unprofessional” manner. Likewise, Mr. Harvey told the grand jury that he had “us[ed] choice language . . . [with Mr. Moore] because of the way [Mr. Moore] was describing” the prosecutor. No doubt this is why the majority throws in “outrage” and its vague catchall “other strong emotion” to the possible reactions that might render an attorney-client communication an abuse. Ante majority at 20, 23. But to say that any conversation in which an attorney is offended (even strongly so) by what their client says to them abuses the relationship underestimates the attorneys who engage in direct client representation and undermines the services they provide. It robs counsel of their professionalism (which Mr. Harvey concededly abandoned), and it empowers counsel, motivated by anger, to weaponize confidential information against their clients. The trial court’s reaction to Mr. Harvey’s motion to withdraw after he revealed Mr. Moore’s statements underscores that, in the real world, the mere fact 104 that Mr. Moore had made statements that could be perceived as threats to the prosecutor did not mean that the attorney-client relationship had been abused. The court acknowledged that “this case ha[d] been a challenge for [Mr. Harvey] and that [his] client c[ould] be somewhat difficult,” but in its view the situation did not “necessarily rise[] to that point where [Mr. Harvey] couldn’t continue to represent him in this case.” The court even relayed her own experience of being threatened by a litigant and explained that “people make threats all the time to get a judge off the case, to get a prosecutor off a case, to get an attorney off a case.” Mr. Harvey then had to engage in an extended discussion, spanning twenty pages of transcript, to persuade the court that he should be permitted to withdraw—not because of Mr. Moore’s conditional statements that he wanted to shoot the prosecutor, but because he and Mr. Moore for various reasons did not have a “healthy relationship” and could no longer communicate effectively. 30 See supra Part III.A. 30 Our colleagues devote considerable attention to a scenario not presented by this case: where a client, seeking legal advice such that the privilege applies, threatens counsel. See ante majority at 23-24. Even if we conceded for the sake of argument that a threat of death or serious bodily injury against the lawyer themself could constitute an abuse of the relationship such that the privilege should not apply because becoming the object of a threat is different in kind from simply hearing one, any hypotheticals envisioning such a circumstance cannot support the majority’s exception in this case. And the majority’s criticism that the dissent is engaged in mere “line-drawing” “not grounded in principle,” id., is surprising given that it is the majority that is carving out a never-before-seen exception to a centuries old common law privilege (defined in relation to the District’s threats statute, but drawing the line 105 Alternatively, the majority posits that uttering a statement to counsel that could be perceived as a threat is an abuse of the relationship because counsel might “feel[] morally obligated to report the threat to law enforcement, who in turn [might] warn the threat’s target,” thereby placing that individual in fear. Ante majority at 21. This construct of “abuse” also does not withstand scrutiny. First, recall that for a communication to be privileged in the first place, it must be “made in confidence,” judged by “the intent of the person seeking advice” and what they “reasonably believed” when they made the statement. Jones, 828 A.2d at 175-76 (first quoting Wigmore, supra § 2292). For the majority’s exception to even be relevant, therefore, the client must have made the statement with the intent and reasonable belief that it would not reach its subject. 31 Recall, as well, that it is a core part of an attorney’s job to be a sounding board, a buffer, and a filter for their client and that they ordinarily must hold their client’s communications in confidence. And if the lawyer were to reveal these statements to law enforcement in this scenario, it would at death or substantial bodily harm) in seeming disregard of the utilitarian and normative principles that undergird the privilege. See supra Part I. 31 The majority responds by saying that a statement is a threat even if it is not intended to reach its target, and that therefore the client’s intent should not affect whether the statement is an abuse either. Ante majority at 22 n.8. But that argument simply assumes that because the statement constitutes a threat, it is an abuse of the relationship—the very question we are debating. The majority’s circular reasoning would seem to render any communication with an attorney that happens to have a negative consequence, whether intended or not, an abuse of the relationship and subject to exception from the privilege. 106 likely be pursuant to ethical Rule 1.6(c) because counsel “reasonably believe[d]” that his client was going to engage in a “criminal act . . . likely to result in death or substantial bodily harm absent disclosure.” That is, the lawyer would be making the disclosure to protect both a third party from experiencing harm and his client from committing harm and exposing himself to legal consequences. 32 Lastly, the lawyer would have no knowledge that law enforcement would necessarily pass a disclosed statement on to the subject. Law enforcement might conclude after investigation that the client did not utter a true threat or did not present any real danger to the subject. And even assuming law enforcement decided to relay the threat to the subject, the subject might be assured that law enforcement was aware of the situation and was either investigating or pursuing charges. 33 Given the above, it is unclear how any possibility of the threatening statements reaching their target would render those statements an abuse of the attorney-client 32 Presumably trial judges do not order lawyers to disclose confidential communications under Rule 1.6(e), as occurred in this case, with regularity. If the majority’s new exception is an invitation to courts to order disclosures first and litigate issues of privilege later as a matter of course, that presents additional problems. 33 Relatedly, setting aside that the privilege would be defeated if the client intended the attorney to disclose a threat, see supra, it would be strange for any client to utter a threat to their lawyer with the intent that the threat would reach its subject in this way. People do not usually desire their criminal threats to be reported to the police. 107 relationship. Instead, the hypothetical the majority posits would, at worst, consist of statements that the client never intended to reach their subject being passed on to law enforcement for the sake of the subject’s safety, and then being disclosed to the subject, also for their safety, with the understanding that law enforcement was handling the situation. Any fear the subject experienced as a result of being advised of the threat would not be the product of intent by any actor to cause the subject any fear; it would be a product of law enforcement doing its job. The above analysis is academic, however, because there is no requirement under the majority’s novel exception that an attorney make any disclosure of their client’s statements before its threats exception applies. And the majority cannot rely on the fact that a disclosure happened to have been ordered in this case; its new exception allows the privilege to be abrogated even when there is no disclosure. One can imagine various scenarios where that might occur: for example, where a communication the attorney and client thought was taking place in private was not. The majority’s justification thus ill fits the exception it creates. B. The Lack of Analogue for the Majority’s Threats Exception The majority’s exception is not only analytically ill-founded and unsupported by the crime-fraud exception, it also has no analogue in any jurisdiction in this country. The dissent conducted our own survey of all fifty states, every federal 108 circuit, Puerto Rico, Guam, and the Virgin Islands. We found no jurisdiction that creates a standalone exception to the attorney-client privilege for criminal threats. To understand just how unprecedented the majority’s threats exception is, one has to put it in context of the national landscape with respect to the attorney-client privilege. Unlike the District, most jurisdictions in this country have codified their attorney-client privilege. 34 Most jurisdictions have also codified their exceptions to the privilege, 35 adopting a largely identical list of exceptions, derived from either the 34 See Ala. R. Evid. 502; Alaska R. Evid. 503; Ariz. Rev. Stat. Ann. §§ 12-2234, 13-4062; Ark. R. Evid. 502; Cal. Evid. Code §§ 950 -962; Colo. Rev. Stat. Ann. § 13-90-107(1)(b); Conn. Code Evid. § 5-2; D.R.E. 502; Fla. Stat. Ann. § 90.502; Ga. Code Ann. § 24-5-501; Haw. Rev. Stat. Ann. § 626-1, R. 503; Idaho R. Evid. 502; Ind. Code Ann. § 34-46-3-1(1); Iowa Code Ann. § 622.10; Kan. Stat. Ann. § 60-426; Ky. R. Evid. 503; La. Code Evid. Ann. art. 506; Me. R. Evid. 502; Md. Code Ann., Cts. & Jud. Proc. § 9-108; Ma. R. Evid. 502; Minn. Stat. Ann. § 595.02(b); Miss. R. Evid. 502; Mo. Ann. Stat. § 491.060(3); Mont. Code Ann. § 26-1-803; Neb. Rev. Stat. § 27-503; Nev. Rev. Stat. §§ 49.035-49.115; N.H. R. Evid. 502; N.J. R. Evid. 504; N.J. Stat. Ann. § 2A:84A-20; N.M. R. Evid. 11-503; N.Y. C.P.L.R. 4503; N.D. R. Evid. 502; Ohio Rev. Code Ann. § 2317.02(A); Okla. Stat. Ann. tit. 12, § 2502; Or. Rev. Stat. §§ 40.225-40.227; 42 Pa. Cons. Stat. Ann. §§ 5916, 5928; P.R. Laws Ap. tit. 32a, § IV, R. 25; S.D. Codified Laws § 19-19- 502; Tenn. Code Ann. § 23-3-105; Tex. R. Evid. 503; Utah R. Evid. 504; Vt. R. Evid. 502; 5 V.I. Code Ann. § 852; Wash. Rev. Code Ann. § 5.60.060(2)(a); Wis. Stat. Ann. § 905.03; Wyo. Stat. Ann. § 1-12-101(a)(i); see also In re Ti.B., 762 A.2d 20, 27-28 (D.C. 2000) (explaining that the attorney-client privilege in the District is a matter of common law). 35 Ala. R. Evid. 502(d); Alaska R. Evid. 503(d); Ark. R. Evid. 502(d); Cal. Evid. Code §§ 956-962; D.R.E. 502(d); Fla. Stat. Ann. § 90.502(4); Haw. Rev. Stat. Ann. § 626-1, R. 503(d); Idaho R. Evid. 502(d); Iowa Code Ann. § 622.10(2), (4); Kan. Stat. Ann. § 60-426(b); Ky. R. Evid. 503(d); La. Code Evid. Ann. art. 506(C); Me. R. Evid. 502(d); Ma. R. Evid. 502(d); Miss. R. Evid. 502(d); Neb. Rev. Stat. 109 draft Federal Rule of Evidence 503 36 (which was never enacted but serves as the basis for many jurisdictions’ attorney-client privilege rules) or the Uniform Rule of Evidence 502 37 (which was drafted by the Uniform Law Commission for use by state legislatures). 38 See 1 Testimonial Privileges § 1:4 (3d ed.). None has codified an § 27-503(4); Nev. Rev. Stat. § 49.115; N.H. R. Evid. 502(d); N.J. R. Evid. 504(2); N.J. Stat. Ann. § 2A:84A-20(2); N.M. R. Evid. 11-503(D); N.Y. C.P.L.R. 4503(b); N.D. R. Evid. 502(d); Ohio Rev. Code Ann. § 2317.02(A); Okla. Stat. Ann. tit. 12, § 2502(D); Or. Rev. Stat. §§ 40.225(4), 40.252; P.R. Laws Ap. tit. 32a, § IV, R. 25(C); S.D. Codified Laws § 19-19-502(d); Tex. R. Evid. 503(d); Utah R. Evid. 504(d); Vt. R. Evid. 502(d); 5 V.I. Code Ann. § 852(d); Wis. Stat. Ann. § 905.03(4). 36 The draft Federal Rule of Evidence 503 contains five exceptions: (1) “Furtherance of crime or fraud”; (2) “Claimants through same deceased client”; (3) “Breach of duty by lawyer or client”; (4) “Document attested by lawyer”; and (5) “Joint clients.” Proposed Fed. R. Evid. 503, 56 F.R.D. 235, 235-37 (1972). 37 Unform Rule of Evidence 502 contains the same five exceptions as the draft Federal Rule of Evidence 503, as well as two additions: (1) an exception for “a communication necessary for a lawyer to defend in a legal proceeding an accusation that the lawyer assisted the client in criminal or fraudulent conduct” and (2) an exception for communications between public officers or agencies and their lawyers under certain circumstances. Nat’l Conf. of Comm’rs on Unif. State L., Uniform Rules of Evidence Act, R. 502 (2005). 38 See Ala. R. Evid. 502(d) (FRE 503); Alaska R. Evid. 503(d) (FRE 503); Ark. R. Evid. 502(d) (URE 502); D.R.E. 502 (URE 502); Fla. Stat. Ann. § 90.502(4) (FRE 503); Idaho R. Evid. 502(d) (FRE 503 plus additional exception for shareholder actions); Kan. Stat. Ann. § 60-426(b) (FRE 503); Ky. R. Evid. 503(d) (FRE 503); La. Code Evid. Ann. art. 506(C) (FRE 503 plus additional exception for information concerning identity of client); Me. R. Evid. 502(d) (URE 502 minus exception for communications necessary to defend against allegations of assisting in crime or fraud); Ma. R. Evid. 502(d) (URE 502 minus exception for communications necessary to defend against allegations of assisting in crime or fraud); Miss. R. Evid. 502(d) (FRE 503); Neb. Rev. Stat. § 27-503(4) (FRE 503); Nev. Rev. Stat. § 49.115 (FRE 503); N.H. R. Evid. 502(d) (FRE 503); N.J. R. Evid. 504(2); N.J. 110 exception to the attorney-client privilege for threats. 39 A review of the jurisdictions where the attorney-client privilege or the exceptions thereto are still a matter of common law bears the same result. 40 Nonetheless, the majority asserts that its new exception has “strong historical footing.” Ante majority at 24. The majority looks to cases it claims have held that the privilege does not apply to “the commission of crimes completed in the presence of the attorney.” Id. at 24; see also id. at 24 n.9. But all this list of cases reveals is that the majority has found some number of decisions where courts have held for various reasons that the attorney-client privilege did not apply to particular threatening statements. See id. at 25 (admitting that the cases it cites “from other jurisdictions do not all mirror the rule we adopt today”). These decisions are Stat. Ann. § 2A:84A-20(2) (FRE 503 minus exceptions for document attested by lawyer and joint clients); N.M. R. Evid. 11-503(D) (FRE 503); N.D. R. Evid. 502(d) (URE 502); Okla. Stat. Ann. tit. 12, § 2502(D) (URE 502); P.R. Laws Ap. tit. 32a, § IV, R. 25 (FRE 503); S.D. Codified Laws § 19-19-502(d) (FRE 503); Tex. R. Evid. 503(d) (FRE 503); Utah R. Evid. 504(d) (FRE 503); Vt. R. Evid. 502(d) (FRE 503); Wis. Stat. Ann. § 905.03(4) (FRE 503); see also Proposed Fed. R. Evid. 503, 56 F.R.D. 235, 235-37 (1972); Nat’l Conf. of Comm’rs on Unif. State L., Uniform Rules of Evidence Act, R. 502 (2005). 39 See supra note 35. 40 The majority’s exception is also an outlier among common law jurisdictions because it creates a common law exception to the privilege based on the statutory definition of a criminal offense. It is unclear why the District’s threats statutes should dictate the contours of a common law privilege, and the majority points to no common law jurisdiction that has rooted an exception to the privilege in the statutory definition of a criminal offense. 111 unremarkable and provide no support for the majority’s novel threats exception. The dissent acknowledges that statements that could be perceived as threats are not unqualifiedly protected by the privilege. If, for example, unlike in this case, a client’s threatening statements did not satisfy the criteria for the privilege (because they were not made in confidence) or if the statements fell under the crime-fraud exception (because the attorney agreed to help the client carry out the threatened conduct), we would readily agree that the privilege did not apply. Needless to say, it is not “miss[ing] the forest for the trees,” id. at 29, to highlight that the majority’s reliance on caselaw from other jurisdictions turns on outcome, not analysis, see id. (admitting it “may stand alone in [its] specific reasoning” and the cases it cites illustrate only that courts have held “one way or another” that particular threats were not protected by the privilege). The majority leads with Coveney v. Tannahill, 1 Hill 33, 41 (N.Y. Sup. Ct. 1841), which it characterizes as “a widely read New York case” from the mid- nineteenth century, ante majority at 24. In Coveney, a case addressing whether a lawyer could be forced to testify about his client’s communications about an alleged fraud, the court explained that a client’s communications are protected by the attorney-client privilege, but “the acts done by him in the presence of the attorney” are not. Coveney, 1 Hill at 35. The court declined to “undertake to say how far the distinction between the communications and the acts of the client may extend.” Id. 112 at 36. But the court cautioned that “[i]f any communications passed between the attorney and client apart from [a third party], these may be privileged.” Id. at 37. And it held that the attorney could not be required to answer a question about the content of a document that was part of the allegedly fraudulent transaction because “all confidential communications between attorney and client, whether written or oral, are alike privileged.” Id. at 42. Coveney is thus a far cry from the strong historical support the majority makes it out to be for its novel threats exception. The majority may believe that uttering a threat to an attorney is more akin to an action taken in an attorney’s presence and less akin to communications we traditionally consider to be privileged. But Coveney does not speak to that proposition. Apart from Coveney, the majority also cites a number of cases in a footnote. See ante majority at 24 n.9. But these cases support neither the majority’s proposition that courts generally “exclude[] from the attorney-client privilege the commission of crimes completed in the presence of the attorney,” nor its threats exception. Id. at 24. In two cases cited by the majority, the courts held that a specific threat was not privileged because it did not meet the test for a privileged communication in the first place. See id. at 24 n.9 (first citing State v. Mewherter, 46 Iowa 88 (1877), and 113 then citing United States v. Alexander, 287 F.3d 811 (9th Cir. 2002)). 41 But whether some threats might not constitute privileged communications under the test we already have—again, a proposition the dissent does not contest—says little about whether our court should create a new exception for those threats that do constitute privileged communications under that test. In another case, United States v. Ivers, 967 F.3d 709 (8th Cir. 2020), the court held that all threats by their very nature fail the test for privileged communications. Id. at 716. But earlier in its opinion, the majority rightly “decline[d] to rely on a conclusion that serious, objectively credible threats do not meet the elements of the attorney-client privilege—for example, because they necessarily are not related to the legal representation or do not seek legal advice.” Ante majority at 11 (footnote omitted). And so Ivers—which has been criticized by commentators, see, e.g., Rice, supra § 5:21, and which the concurrence describes as “reflect[ing] too impoverished a view of the attorney-client privilege,” ante Deahl, J., concurring at 54; see also supra note 21—does not support the majority’s creation of a new exception to the privilege either. 41 Note that Iowa has since codified its attorney-client privilege and has not created an exception for threats. See Iowa Code Ann. § 622.10. 114 The majority also cites two cases discussing an exception to the privilege for communications that indicate a reasonable probability or intent to commit future harm or future criminal conduct. See ante majority 24 n.9 (first citing State v. Hansen, 862 P.2d 117 (Wash. 1993) (en banc), and then citing People v. Dang, 113 Cal. Rptr. 2d 763 (Ct. App. 2001)). But, again, an imminent harm exception is conceptually different from and arguably narrower than the majority’s exception for “completed criminal threats,” ante majority at 30, which need not presage some actual future harm, see id. at 15 (explaining “one need not intend to follow through on the threat for it to be criminal”). Lastly, the majority cites a grab bag of cases for quotes that are taken out of context and reflect neither the holding of the case nor the present rule in that jurisdiction. For example, the majority quotes the court’s pronouncement in McMannus v. State, 39 Tenn. 213 (1858), that “[i]t would be monstrous to hold, that if counsel was asked and obtained, in reference to a contemplated crime, that the lips of the attorney would be sealed, when the fact might become important to the ends of justice in the prosecution of crime.” Id. at 216; see ante majority at 24 n.9. But that case concerned a defendant who had asked for legal advice from someone he had just met and who he did not necessarily know was a lawyer; read in context, this quote simply explains that it would make no sense to apply the privilege in that scenario. See McMannus, 39 Tenn. at 214-16. 115 The majority also quotes the court’s observation in Hopkinson v. State, 632 P.2d 79 (Wyo. 1981), that: “We cannot imagine a threat of injury made by a client toward the family and property of an attorney as being privileged and within those communications protected.” Id. at 116; see ante majority at 24 n.9. But the question on appeal in that case was whether the government had erred by referring to a threat, which the trial court later deemed inadmissible, in its opening statement at trial. Hopkinson, 632 P.2d at 115. The appellate court held that the reference to the threat was not error because the government could have believed in good faith that the trial court would have allowed the testimony to come in, for what reason the appellate court did not say. Id. at 115-16. The appellate court did not explain whether it thought the threats to counsel would in fact be admissible because they did not satisfy the criteria for the privilege, because they would fall under some exception to the privilege, or for some other reason. See id. at 115-16. No subsequent case citing Hopkinson has understood it to have created a threats exception to the attorney-client privilege or even quoted this portion of the opinion. The majority quotes Jackson v. State, 293 S.W. 539 (Tenn. 1927), as saying that “threats made by a client against the life of a person during a professional consultation with his attorney are not privileged.” Id. at 540 (quoting 28 Ruling Case Law 559, 560); see ante majority at 24 n.9. But the court in Jackson simply held that the statement at issue was not privileged because it had nothing to do with 116 the matter about which the client was consulting the attorney. See Jackson, 293 S.W. at 540. There is no subsequent case suggesting that there is a threats exception in Tennessee, and certainly not one citing Jackson for that proposition. And the majority cites State v. Nixon, 207 P. 854 (Kan. 1922), characterizing it as quoting with approval the trial court’s statement: “I don’t think anybody would claim that, if the defendant had gone into [the attorney’s] office, and said to him, ‘I am going to kill [the victim],’ and then consulted with him, it would claim such a matter was privileged.” Ante majority at 24 n.9 (alterations in original) (quoting Nixon, 207 P. at 855). But as the (italicized) text in the full quote reproduced in the Kansas appellate court’s opinion makes clear, the trial court was simply giving an example of a client making a statement entirely unrelated to the topic about which they were consulting the lawyer: I don’t think that a man may come into a lawyer’s office and talk about matters that have nothing whatever to do with his business that he is there upon, and come in and claim that it is a confidential communication, so that the attorney cannot testify. In this case I think this evidence was just as competent, as I don’t think anybody would claim that, if the defendant had gone into [the attorney’s] office, and said to him, ‘I am going to kill [the victim],’ and then consulted with him, it would claim such a matter was privileged . . . . Nixon, 207 P. at 855 (emphasis added). As explained above, the dissent does not disagree that some threats may be unprivileged because they do not meet the existing 117 test for privileged communications, but that in no way supports the majority’s categorical exception. 42 42 The majority dismisses cases cited by Mr. Moore and PDS as amicus where threatening statements have been deemed protected by the attorney-client privilege—see, e.g., Purcell v. District Attorney, 676 N.E.2d 436, 438, 441 (Mass. 1997); Newman v. State, 863 A.2d 321, 330-36 (Md. 2004); In re Grand Jury Investigation, 902 N.E.2d 929, 930-34 (Mass. 2009); State v. Boatwright, 401 P.3d 657, 660-61, 664 (Kan. Ct. App. 2017)—as reflecting a “minority approach,” ante majority at 28. But this characterization is premised on the majority’s failure to acknowledge the dissent’s jurisdictional survey and its overselling of its collection of largely irrelevant authority. The majority also tries to undermine the persuasive authority of these cases, but with little success. The majority incorrectly claims that Purcell and Newman are factually distinguishable. The majority asserts that Purcell did not “involve[] a criminal threat,” ante majority at 26 n.10, but it did, Purcell, 676 N.E.2d at 438 (describing the communication at issue as one in which the client “had made threats to burn the apartment building”). If the majority is relying on the fact that the statement was admitted in an arson prosecution, we do not see the relevance of that fact. We understand the majority to be creating an exception to the privilege that allows in any type of case, criminal or civil, the admission of any statement to an attorney in the course of seeking legal advice that could be deemed a completed criminal threat. And in any event, Massachusetts has since applied the rule in Purcell as binding in a criminal threats prosecution. See In re Grand Jury Investigation, 902 N.E.2d 929, 930, 932-34 (Mass 2009). As for Newman, the majority states that it involved “the agreement portion of a conspiracy,” ante majority at 26 n.10, but nowhere does the court’s analysis rely on the fact that the underlying case was one for conspiracy, see Newman, 863 A.2d at 298-333, 335-37. As for In re Grand Jury and Boatwright, the majority asserts these decisions are unpersuasive because neither addresses its concern that a threat uttered to counsel in the course of seeking legal advice is an abuse of the attorney-client relationship. See ante majority at 26-28. But the same critique could be made of the cases the majority cites in its favor. Only the majority espouses the idiosyncratic view that uttering a statement to counsel that meets the test for privileged communications but could be considered a criminal threat constitutes an abuse of the attorney-client relationship. In short, the majority’s critique just reveals the novelty of its reasoning. 118 That the majority and the dissent have been unable to find a single jurisdiction that has created the exception the majority creates with this opinion firmly refutes the majority’s claim that it breaks “little new ground.” 43 Ante majority at 26. In all the centuries that the attorney-client privilege has existed in the United States, no jurisdiction has ever considered, much less endorsed, the threats exception that the majority creates. 43 The majority argues that not creating this new exception is what would truly be unprecedented because it would “in effect be creating an attorney-client exception to the criminal threats statute.” Ante majority at 29 n.11. This is untrue for two reasons. First, the privilege might not apply to the client’s threat, as the majority’s own cases reveal occurs with regularity. See ante majority at 24 n.9; see also id. at 38 (“Most threats—even those uttered in the presence of an attorney—are presumably not in pursuit of legal advice and thus lack privilege either way.”). Second, even if the privilege did apply, this would simply be yet another instance, often seen in Superior Court, where the immediate recipient of a threat is unavailable to testify. The government could still prosecute the alleged threat using whatever other evidence was available and admissible, perhaps the testimony of an unknown overhearer or (as in this case) surveillance video. What the majority really seems to be saying is that, without its new exception, it will be harder to convict people in criminal threats cases. But recall that privileges by their nature make it harder to reach the “correct” result in court, and the existence of the privilege in the first place rests on our belief that the protection of the attorney- client relationship and the proper functioning of our adversarial system is more important than the need to prosecute every criminal offense, or reach the correct result in other types of cases. See supra Part I. The need to prosecute more threats cases therefore is not a proper consideration in creating a novel exception to the attorney-client privilege. 119 C. The Consequences of the Majority’s Threats Exception The majority’s new threats exception is not just ill-conceived and unprecedented. It will have far-reaching consequences for the contours and application of the privilege. Although the majority presents the exception as a modest carveout from otherwise privileged communications, it is anything but. As a consequence, its chilling effect will be significant. 1. The extent of the intrusion The majority presents its new threats exception as a limited incursion on otherwise-privileged conversations. But in concept, the majority’s new threats exception is a significant intrusion, and the limitation the majority imposes on its exception is, in fact, no limitation at all. First, as discussed, this exception requires nothing more of counsel than that they do their normal job of listening to their client. Counsel need not disclose their client’s statements (as permitted under limited circumstances under the ethical rules), nor must counsel transmit the client’s statements to anyone. Second, as for what the client says to counsel, the exception does not just apply to completed “criminal threats.” It casts a far wider net. The elements of threats need not be proven beyond a reasonable doubt before the privilege is lifted 120 and may never be so proven. The majority speaks of judges making a “preliminar[y]” assessment of the alleged statement to determine whether the exception applies. Ante majority at 40. If that preliminary showing is analogous to the standard set forth in In re PDS, then communications may be deemed by a judge to be unprivileged upon a mere showing of probable cause, see In re PDS, 831 A.2d at 904, and a jury could then determine that the government had failed to prove beyond a reasonable doubt that the communications were threats. 44 Or the communications could be compelled in civil discovery on a mere showing of relevance and used in a case where there would never be a determination beyond a reasonable doubt that the communications constituted criminal threats. See Super. Ct. Civ. R. 26(b)(1) (authorizing discovery of “any nonprivileged matter that is relevant to any party’s claim or defense”). Moreover, as a practical matter, courts will likely deem any objectively credible statement to also satisfy the mens rea requirement of threats under this lower standard of proof. In short, the majority’s exception is not actually for proven criminal threats, but for any communications that could be deemed to possibly constitute a threat and will include communications that are not ultimately proven to be threats. 44 In re PDS grounded its probable cause standard in the fact that testimony was being sought before a grand jury, see 831 A.2d at 904, but it is unclear whether the same standard would suffice to deem testimony unprivileged in trial proceedings. 121 Third, this exception is not tailored to any real-time or even future public safety objective. This exception applies whether or not anyone—the attorney who hears the client’s statement, the government or private attorney who seeks to invoke the exception, or the judge who rules on the request to abrogate the privilege— actually believes that the client intended to carry out the threat. See ante majority at 15. Thus, even in a situation where the attorney concludes that the client had communicated an “objectively credible” threat in the course of seeking legal advice, but thereafter has a productive conversation with the counsel that provides assurance that the client will conform their conduct to the law going forward, the privilege will be abrogated. 45 Or, even where the attorney never believes the client communicated 45 The majority justifies this broad application of its threats exception by reasoning that “the crime of a criminal threat has already been committed,” “the lawyer at least has already suffered the harms associated by that crime,” and to focus “on the mere possibility that the lawyer might prevent, through the power of reasoning, a future crime both underestimates the harm that threats cause and overestimates attorneys’ powers of persuasion.” Ante majority at 39. Again, we think the majority’s conception of “harms,” in the form of possible adverse emotions from hearing a client utter a threat against a third party, ignores the realities of direct client representation. See supra Part IV.A. On the other side of the equation, we stand by prioritizing the prevention of actual death or serious bodily injury over the prosecution of threatening words. More generally, we defend the logic of the privilege, which is based on the understanding that an attorney’s “powers of persuasion,” ante majority at 39, are “indispensable” to “our administration of justice,” In re PDS, 831 A.2d at 900 (quoting Hickman v. Taylor, 329 U.S. 495, 515 (1947)). As we explained in In re PDS: A lawyer’s advice is important when clients honestly seek guidance as to their duties under complicated regulatory 122 an objectively credible threat or intended to carry it out, if the other party makes a preliminary showing to the judge that the client uttered an objectively credible threat, the privilege will also be abrogated. Moreover, this exception does not just allow admission of statements that could be construed as threats in criminal prosecutions; it allows for the admission of these statements in cases of any kind, including cases that have nothing to do with keeping any individual or the public safe. There is also no time limitation; the privilege may be abrogated months or even years after the statements took place. 46 regimes, but it is even more vital when the client misguidedly contemplates or proposes actions that the client knows to be illegal. The existence of the attorney- client privilege encourages clients to make such unguarded and ill-advised suggestions to their lawyers. The lawyer is then obliged, in the interests of justice and the client’s own long-term best interests, to urge the client, as forcefully and emphatically as necessary, to abandon illegal conduct or plans. The sincere counsel of a trusted advisor will persuade many clients to comply with the law. Indeed, discouraging clients from illegal conduct is a regular occurrence in an attorney’s practice. . . . Thus, when clients contemplate presenting false evidence, lawyers are the first and most effective line of defense for the integrity of the judicial process. Id. at 900-01 (internal citations omitted). Although the majority opinion appears to question this line of thinking, between Judge Deahl’s concurrence and this dissent, a majority of this en banc court reaffirms “the entirety of this court’s analysis in In re PDS.” Ante Deahl, J., concurring at 60-61. 46 The majority does not respond to this point other than to assert that mere disclosure by counsel (presumably under Rule 1.6(c), which did not occur here) is 123 The majority’s global response to these concerns about the breadth of its newfound exception—namely, that the same critiques could be leveled against the crime-fraud exception, ante majority at 33—is unpersuasive. The majority fails to grapple with the unique difficulty of discerning true criminal threats from angry venting. The majority also ignores the fact that the crime-fraud exception has a built- in requirement that raises the bar for when it applies; as noted above, it actually requires counsel to assist their client’s criminal scheme in some way, see supra Part III.C., unlike the majority’s new threats exception, which requires nothing more of counsel than to serve as passive hearer and dismisses the possibility that counsel might dissuade their client from acting on the threat. In short, the majority fails to consider its novel threats exception on its own terms. Lastly, the limitation that the majority imposes (at least for now) on its exception—requiring that the threat relate to death or serious bodily injury—does not address these concerns about its breadth. It is not a real limitation at all, because the majority simply “leave[s] for another day whether to exempt other crimes and lesser threats from the attorney-client privilege.” Ante majority at 32. not enough to “improve public safety”; it must be followed by arrest and prosecution. Ante at majority at 41. But again, the majority does not actually require disclosure by counsel, and its argument is based on an unsubstantiated assumption that arrest will occur in real time, the client will be incarcerated pretrial, and successful prosecution will follow soon thereafter. 124 2. Chilling attorney-client communications The majority nowhere acknowledges that the risk of chilling legitimate attorney-client communications is an essential consideration in determining whether to create a brand-new exception to the privilege. Instead, it simply denies that any meaningful chilling will occur. But in so doing it woefully underestimates the danger its exception poses. The Supreme Court and this court have at various times explained that a person may react to rules governing speech by self-censoring even permitted speech, and we thus “have been especially anxious to assure to the freedoms of speech and press th[e] ‘breathing space’ essential to their fruitful exercise.” Gertz v. Robert Welch, Inc., 418 U.S. 323, 342 (1974) (quoting NAACP v. Button, 371 U.S. 415, 433 (1963)); see also N.Y. Times Co. v. Sullivan, 376 U.S. 254, 276 (1964); accord Nader v. de Toledano, 408 A.2d 31, 43 (D.C. 1979). The Supreme Court discussed this chilling effect in the context of threats recently in Counterman v. Colorado, 600 U.S. 66 (2023). The Court explained that “[p]rohibitions on speech have the potential to chill, or deter, speech outside their boundaries” for multiple reasons: (1) “A speaker may be unsure about the side of a line on which his speech falls”; (2) “he may worry that the legal system will err, and count speech that is permissible as instead not”; or (3) “he may simply be concerned about the expense of becoming entangled in the 125 legal system.” Id. at 75. To avoid the “result [of] ‘self-censorship’ of speech that could not be proscribed,” the court in Counterman endorsed a heightened mens rea requirement, even as it recognized this requirement “will shield some otherwise proscribable (here, threatening) speech.” Id. But the Court determined the cost was worth it because “the added element reduces the prospect of chilling fully protected expression,” or “said a bit differently,” it “provides breathing room for more valuable speech.” Id. (alterations and quotation marks omitted). In the attorney-client privilege context just as in the First Amendment context, “[t]his Court again must consider the prospect of chilling non-threatening expression, given the ordinary citizen’s predictable tendency to steer ‘wide[] of the unlawful zone’” out of a “fear of mistaking whether a statement is a threat,” “fear of the legal system getting that judgment wrong,” and “fear, in any event, of incurring legal costs—all [of which] may lead him to swallow words that are in fact not true threats.” Id. at 77-78 (emphasis added) (alteration in original) (quoting Speiser v. Randall, 357 U.S. 513, 526 (1958)); see also Imwinkelried, supra § 3.2.3(2) (“To borrow an expression from the case law on the First Amendment overbreadth doctrine, the assumption is that the normal person’s fear of compelled public disclosure would have a ‘chilling effect’ on the person’s willingness to confer and communicate.” (footnotes omitted)); cf. Jaffee v. Redmond, 518 U.S. 1, 11-12 (1996) 126 (“If the [psychotherapist-patient] privilege were rejected, confidential conversations between psychotherapists and their patients would surely be chilled . . . .”). It is notable that the only party before the en banc court that actually represents individual clients, the Public Defender Service, warns us that carving out a threats exception to the attorney-client privilege will have a chilling effect on desirable attorney-client communications. The majority has three responses. Ante majority at 34-36. First, the majority asserts that “clients are presumably used to communicating without pronouncing objectively credible threats because the prohibition on issuing criminal threats applies to our day-to-day conversations as well,” effectively arguing people should know better. Id. at 34-35. But as discussed, the majority ignores the strong likelihood that when individuals consult with their lawyers, they are not speaking about day-to-day events and may be confronting the most stressful circumstances of their lives. See supra note 28. These consultations deserve a bit more breathing room. Moreover, the majority ignores other reasons chilling occurs: the boundaries of what constitutes a threat may not be clear, and even if people are confident in their own abilities not to cross the line, they may suppress even protected and valuable speech for fear that the legal system will err or that, even if ultimately vindicated, they will have to incur the expense and stigma of being embroiled in the legal 127 system. See Counterman, 600 U.S. at 75. This concern is particularly salient where, as here, the elements of a threat need not be proven beyond a reasonable doubt for the communications to be deemed unprivileged. Second, the majority asserts that there should be no concern about chilling because “excluding criminal threats from the privilege . . . is hardly a new development.” Ante majority at 35. But as discussed, the majority points to only a handful of cases, none of which endorse an analogous exception to the privilege for threats. See supra Part IV.B. The majority also says that “we have never before suggested that criminal threats would be privileged,” ante majority at 35, but we have never suggested that statements that could be construed as threats confidentially made to counsel in the course of seeking legal advice would not be privileged. Third, the majority claims that any chilling from its threats exception will be “minimal” because counsel “can already disclose serious, objectively credible threats made by their clients,” apparently referring to Rule of Professional Conduct 1.6(c). Id. at 35. But disclosure under that rule is only permissible where counsel believes that the client will carry out the threatened action, whereas the majority’s exception covers all completed threats whether or not the attorney believes the client intends to carry them out. Further, as noted, the majority opinion 128 does not require a prior disclosure under Rule 1.6(c) for its exception to apply (it could not, else it would have to reverse Mr. Moore’s conviction, see supra Part III.A). 47 And there is a significant difference between making a limited disclosure to avert an immediate harm and taking the stand to testify against one’s client. Cf. Chase, 340 F.3d at 988 (explaining that “a patient will retain significantly greater residual trust when the therapist can disclose [a threat of harm to another] only for protective, rather than punitive, purposes. In other words, the marginal additional harm to the relationship [from abrogation of the federal psychotherapist-patient testimonial privilege] is significant”); 48 United States v. Hayes, 227 F.3d 578, 584- 85 (6th Cir. 2000) (“While early advice to the patient that, in the event of the disclosure of a serious threat of harm to an identifiable victim, the therapist will have a duty to protect the intended victim, may have a marginal effect on a patient’s candor in therapy sessions, an additional warning that the patient's statements may 47 The majority also “doubt[s] much if any of the relationship between the attorney and the client will survive” after the attorney discloses a threat. Ante majority at 36. But it is unclear why it would not, particularly if the disclosure is under Rule 1.6(c) and if the client comes to see that their attorney acted in their best interest. And even in this case, where the disclosure was made under Rule 1.6(e), the court did not perceive this event as a basis to grant his motion to withdraw, and Mr. Moore opposed Mr. Harvey’s motion. 48 To support its opposing view, the majority cites the concurrence in Chase, which dissented on this particular point. See ante majority at 35-36; see also Chase, 340 F.3d at 993, 996-97 (Kleinfeld, J., concurring in the result). 129 be used against him in a subsequent criminal prosecution would certainly chill and very likely terminate open dialogue.”). * * * The attorney-client privilege has served as a foundation of our adversarial justice system for centuries precisely because it has functional utility and normative value. Until now, no jurisdiction has ever determined that a categorical threats exception to statements that otherwise fall within the privilege’s exception is needed. The dissent disagrees with the majority that Mr. Moore’s case demonstrates that our jurisdiction needs such an exception now. We would heed the Supreme Court’s admonition not to endorse a “no harm in one more exception rationale [that] could contribute to the general erosion of the privilege, without reference to common-law principles of reason and experience.” Swidler & Berlin, 524 U.S. at 410 (quotation marks omitted). For these reasons, we respectfully dissent. Appendix Division opinion issued on Nov. 17, 2022, Part I, reinstated per en banc opinion issued on Sept. 4, 2025 (Moore v. United States, 285 A.3d 228 (D.C. 2022) (Part I)), rest of opinion vacated by Moore v. United States, No. 19-CF-0687, 2023 WL 3674377, at *1 (D.C. May 25, 2023) (per curiam). I. Sufficiency A. Trial Facts and Procedural History At Mr. Moore’s May 2019 jury trial for threatening a public official and obstructing justice, the government called three witnesses: the District of Columbia AAG who had prosecuted Mr. Moore for contempt, a Deputy United States Marshal assigned to investigate the threats against the AAG, and Mr. Harvey. Because Mr. Harvey was the only witness who actually heard what Mr. Moore said, the government’s case rested on his testimony. Mr. Harvey, a longtime member of the Superior Court’s Criminal Justice Act panel, 1 explained that he heard Mr. Moore’s statements because he was Mr. Moore’s court-appointed lawyer in the contempt case. Mr. Harvey testified that the 1 See Criminal Justice Act (CJA) Attorneys, District of Columbia Courts, https://www.dccourts.gov/services/cja-practitioner; https://perma.cc/SUG6-DHLU (last visited Nov. 15, 2022). 2 statements in question were made on two occasions during Mr. Moore’s 2018 contempt trial, which spanned several months so that Mr. Moore, who was not detained and lived in North Carolina, would not miss too many consecutive days of work. On both occasions, the statements were made after the AAG sought to alter Mr. Moore’s conditions of release. Prior to the first incident on April 12, 2018, the AAG asked the court to reverse its order discontinuing GPS monitoring of Mr. Moore via an ankle bracelet. Mr. Harvey and Mr. Moore met in the hallway outside the courtroom to discuss this development, or more particularly, Mr. Moore’s feelings about this development. Mr. Moore was “very agitated” and began by saying things like “[f]uck that bitch. I hate this bitch,” referring to the AAG. Responding to Mr. Moore, Mr. Harvey explained that the AAG was doing her job as a prosecutor, and it was “just silly on his part to be angry.” This only further angered Mr. Moore, who not only repeated “fuck that bitch” but also added “I’ll shoot that bitch.” When Mr. Harvey said, “Man, what are you talking about?” Mr. Moore replied, “That’s right, Harvey. I’ll shoot that bitch.” Mr. Harvey told Mr. Moore he was “starting to . . . think [Mr. Moore was] serious,” prompting Mr. Moore to say, “God damn right, Harvey. Fuck that bitch. I’ll shoot that bitch.” Mr. Harvey then told Mr. Moore he would have to withdraw from representing him and left to call Bar Counsel. 3 Mr. Harvey testified that he called Bar Counsel to “find out what [his] options were.” Mr. Harvey explained that he was aware that, within the scope of his representation of Mr. Moore on contempt charges, he could not disclose Mr. Moore’s “secrets” about past criminal activity, but Mr. Harvey’s understanding was that Mr. Moore’s threats to commit future criminal activity fell outside that representation. 2 Mr. Harvey also noted that under Rule 1.6 of the District of Columbia’s Rules of Professional Conduct, he was authorized to “reveal client confidences and secrets[] to the extent reasonably necessary . . . to prevent” a client from engaging in a criminal act that he as the attorney “reasonably believe[d] [wa]s likely to result in death or substantial bodily harm absent disclosure.” See D.C. R. Prof. Conduct 1.6(c)(1). Mr. Harvey testified that Bar Counsel advised him that the decision whether to disclose such statements under this rule was left to his discretion. Relying on a different rule, D.C. R. Prof. Conduct 1.16 (regarding declining or terminating representation), Mr. Harvey decided to ask the court if he could withdraw because of an unspecified “ethical problem.” He did not indicate which 2 Mr. Harvey explained to the jury, “If [a client] says, ‘I’m going to shoot the President,’ that means [they are] going to do something in the future. I cannot participate in that because, if I give [them] any legal advice, then, basically, I’m helping [them] and guiding [them] through the process to commit a crime as opposed to advising [them] about what the legal ramifications are because [they] committed one.” 4 subsection of Rule 1.16 he wished to rely upon. When the judge asked if Mr. Harvey reasonably believed that Mr. Moore had used or was attempting to use his services to perpetrate a crime or a fraud, alluding to D.C. R. Prof. Conduct 1.16(b)(1) and (2), Mr. Harvey told the court he had “concerns.” But Mr. Harvey later told the judge that his “reason for wishing to withdraw ha[d] nothing to do with [Mr. Moore] requesting [Mr. Harvey] to do anything.” His relationship with Mr. Moore had become “toxic” by this point and Mr. Harvey just “wanted to get out of the case.” The trial court refused to allow him to withdraw based on the information he provided. In the meantime, Mr. Moore informed Mr. Harvey that he had just been “bullshitting” and reassured him, “I didn’t mean it. I didn’t mean it.” 3 Mr. Harvey testified that he told Mr. Moore that he would continue to represent him but instructed him, “You will never, ever use this kind of language with me about anybody because, from this point forward, I’m going to believe you. So if you decide you want to go shoot somebody, you need to keep that to yourself and don’t make me a part of it.” 4 Mr. Moore responded: “All right, Harvey. I’m not According to the Deputy Marshal’s notes from his interview of Mr. Harvey, 3 Mr. Moore also apologized and told Mr. Harvey he was “just blowing off steam.” 4 Before the grand jury, Mr. Harvey additionally testified that he warned Mr. Moore that if Mr. Moore ever threatened the AAG again, he would tell the court. The transcript of Mr. Harvey’s trial testimony on cross-examination is less clear about whether Mr. Harvey warned Mr. Moore only that he would withdraw, or if 5 going—I won’t say nothing like that again. I was just bullshitting.” Mr. Harvey and Mr. Moore then entered the courtroom and the trial resumed. Mr. Moore failed to appear at his next scheduled trial date in June 2018. Mr. Harvey explained to the court that Mr. Moore was training for a new job as a crane operator and had to attend a certain number of classes within a period of time. The judge continued the trial until June 29, 2018, and informed Mr. Harvey that if Mr. Moore did not appear on that date, she would issue a bench warrant for Mr. Moore’s arrest. On June 29, a Friday, Mr. Moore returned to court and the AAG again asked the court to either detain him or order him to resume wearing an ankle bracelet. The court granted the AAG’s request to put Mr. Moore back on GPS monitoring. Because it was nearing 5 p.m., it was too late for Mr. Moore to get an ankle bracelet that day; he had to return to the courthouse on Monday morning. Mr. Harvey testified that Mr. Moore was “pissed off” about this development because it meant that he would have to miss job training scheduled for Monday. Mr. Harvey successfully advised Mr. Moore to keep his anger in check while they were before the judge, but once they were out in the hallway, Mr. Moore expressed his feelings to Mr. Harvey. Mr. Harvey explained that Mr. Moore “wasn’t hollering and Mr. Harvey explicitly told Mr. Moore that he would tell the court what Mr. Moore said. 6 screaming, but you could see [from] the expression on his face that he was . . . mad.” According to Mr. Harvey, they had the following exchange: Mr. Moore: [I]f I lose my job, I’m going to bust a cap in this bitch, I’m going to bust a cap in this bitch. Mr. Harvey: Man, what are you doing? Mr. Moore: Man, fuck this bitch. If I lose my job, I’m going to bust a cap in this bitch [making a hand gesture simulating a gun]. Mr. Harvey: I told you what I was going to do if you ever said something like that to me again. 5 Mr. Moore: Fuck her. Fuck you. Mr. Harvey testified he had “no idea what this man was going to do.” Without further discussion, Mr. Harvey went back into the courtroom and renewed his motion to withdraw. He also told the court that he would reveal Mr. Moore’s statements to him if the court ordered him to, which the court did. After hearing Mr. Harvey’s account of Mr. Moore’s comments, the court immediately ordered Mr. Moore to be detained and subsequently granted Mr. Harvey’s withdrawal motion. 6 5 But see supra note 4. 6 On cross-examination, Mr. Harvey acknowledged that he also told the court that his relationship with Mr. Moore was “horrible” and “[u]nhealthy,” that their discussions had become “unprofessional” because of the profane language both men used with each other, and that he had “never wanted to remove [him]self from a situation as badly as [he] wanted to remove [him]self from this one.” 7 In addition to this testimony from Mr. Harvey and its two other witnesses, the government introduced into evidence silent surveillance footage from June 29, 2018, capturing the second of the two conversations in the courthouse corridor when Mr. Harvey claimed Mr. Moore had threatened the AAG, as well as still photographs taken from the video footage. Mr. Harvey’s grand jury testimony was admitted into evidence in the defense case. Mr. Moore did not testify. The jury quickly convicted Mr. Moore on all counts of threatening a public official and obstructing justice. B. Analysis On appeal, Mr. Moore argues that the government failed to prove that he had the requisite mens rea to support either of his convictions. Although he briefed this argument after his evidentiary claim, we address it first for two reasons: (1) “the Double Jeopardy Clause forbids a second trial for the purpose of affording the prosecution another opportunity to supply evidence which it failed to muster in the first proceeding,” Gray v. United States, 155 A.3d 377, 389 n.21 (D.C. 2017) (internal quotation marks and brackets omitted), meaning Mr. Moore’s evidentiary claim would be moot should he prevail on his sufficiency claim, and (2) “even improperly admitted evidence may be considered in evaluating the sufficiency of the evidence,” Mitchell v. United States, 985 A.2d 1125, 1134-35 (D.C. 2009) (internal 8 quotation marks omitted), meaning that even if we concluded that Mr. Moore’s statements to Mr. Harvey were protected by attorney-client privilege and should not have been admitted at trial, we would have to consider these statements in assessing whether the government proved its case against him. We review sufficiency claims de novo. In re S.W., 45 A.3d 151, 154 (D.C. 2012). “When assessing the sufficiency of the evidence, we view the evidence in the light most favorable to the verdict, giving full play to the right of the fact-finder to determine credibility, weigh the evidence, and draw justifiable inferences of fact.” Miller v. United States, 209 A.3d 75, 77 (D.C. 2019) (internal quotation marks and brackets omitted). “The evidence is sufficient if . . . any rational trier of fact could have found the essential elements of the crime beyond a reasonable doubt.” Id. 1. Threats Against Public Officials In a two-paragraph argument, Mr. Moore asserts the government’s evidence was insufficient to prove that he threatened a public official under D.C. Code § 22-851(c) (providing a criminal penalty for “[a] person who stalks, threatens, assaults, kidnaps, or injures any official or employee . . . while the official or employee is engaged in the performance of his or her duties or on account of the 9 performance of those duties”) because the government did not prove that he intended for his statements to reach the AAG who prosecuted him for contempt. Our analysis is limited to this argument. See, e.g., Sutton v. United States, 988 A.2d 478, 483 (D.C. 2010) (considering sufficiency of the evidence for only the elements of the crime challenged by the defendant). Mr. Moore provides no support for the argument that proof of such a mens rea element is required to establish guilt under D.C. Code § 22-851(c). He cites to only one case from this jurisdiction, Carrell v. United States, 165 A.3d 314 (D.C. 2017) (en banc), and appears to assert that this court implicitly held that, in any threats prosecution, “the government must prove intent for the statement (the conduct) to reach the target (the result)” because “without that, the statement cannot be intended to threaten the target.” 7 Mr. Moore’s reliance on Carrell is misplaced. Sitting en banc, this court reaffirmed in Carrell that the actus reus elements of felony and misdemeanor threats under D.C. Code § 22-407 and § 22-1810 are that the defendant “(1) uttered words 7 Mr. Moore also cites to an intermediate appellate court decision from Massachusetts, Commonwealth v. Maiden, 810 N.E.2d 1279, 1281 (Mass. App. Ct. 2004). He provides no explanation as to how that out-of-jurisdiction decision informs our understanding of D.C. Code § 22-851(c). See id. 10 to another person (2) with a result that the ordinary hearer would reasonably . . . believe that the threatened harm would take place.” 165 A.3d at 320 (internal brackets, quotation marks, and footnote omitted). We also made clear that “the government must prove the defendant’s mens rea to utter the words as a threat, and that it may do so by establishing that the defendant acted with the purpose to threaten or with knowledge that his words would be perceived as a threat.” Id. at 317. But we did not address whether (much less hold that) the government must prove that the defendant’s purpose was for the threat to reach the target, rather than some other person. Certainly we agree that the offense of threats to public officials—like felony and misdemeanor threats—does not criminalize thoughts in one’s head and therefore requires proof that the defendant communicated a threatening statement to someone with the intent (in the sense of purpose or knowledge) that the statement will be perceived as a threat. 8 But, in the absence of any substantiated argument as to why 8 As the government notes, we have held that the crime of threatening under § 22-1810 “[is] complete as soon as the threat [is] communicated to a third party, regardless of whether the intended victim ever [knows] of the plot.” Beard v. United States, 535 A.2d 1373, 1378 (D.C. 1988); see also Gurley v. United States, 308 A.2d 785, 787 (D.C. 1973); Jackson v. District of Columbia., 541 F. Supp. 2d 334, 343 (D.D.C. 2008). But all that means is that the actus reus element of threatening does not require the threat to reach the intended victim (at least under the criminal threats 11 a threat against a public official must be made with the purpose or knowledge that that specific individual will hear the threat, cf. Comford v. United States, 947 A.2d 1181, 1188 (D.C. 2008) (observing that it is generally “not enough merely to mention a possible argument in the most skeletal way, leaving the court to do counsel’s work” (internal quotation marks omitted)), we are unpersuaded that the evidence was insufficient on this basis. 9 2. Obstruction of Justice Mr. Moore also argues that the evidence was insufficient to prove he statute); it does not address whether the defendant must have the purpose or knowledge that the threat will reach the victim. 9 It is undoubtedly the case that, when a threat is delivered only to a third party, that fact can bear on other elements of the offense. For one, it may bear on one’s “mens rea to utter the words as a threat.” Carrell, 165 A.3d at 317; see also, e.g., United States v. Houston, 792 F.3d 663, 665, 667-68 (6th Cir. 2015) (analyzing a conviction under 18 U.S.C. § 875, a federal threats statute, and holding that, when the incarcerated defendant speaking by phone to his girlfriend said, “I’ll kill that motherf[***]er,” referring to his former attorney, a jury could reasonably infer that the defendant was “caught . . . in a fit of rage in a prison cell . . . [and] that he was venting his frustration to a trusted confidante rather than issuing a public death threat to another”). Likewise, it may bear on the actus reus element that an “ordinary hearer would reasonably . . . believe that the threatened harm would take place.” Carrell, 165 A.3d at 320 (internal brackets omitted); cf. Black v. United States, 755 A.2d 1005, 1008 n.7 (D.C. 2000) (holding that a defendant’s words were “of such a nature as to convey fear of serious bodily harm or injury to the ordinary hearer” where he not only stated to a third party that he planned to “put a cap in [the intended victim’s] hand,” but “asked [the third party] to convey the threatening message to [the intended victim]” (internal quotation marks omitted)). But, as noted above, Mr. Moore has not challenged the sufficiency of the evidence on these grounds. 12 obstructed justice under D.C. Code § 22-722(a)(5) (“A person commits the offense of obstruction of justice if that person[] . . . [i]njures or threatens to injure any person . . . on account of the person or any other person performing [their] official duty as a juror, witness, or officer in any court in the District of Columbia . . . .”). Specifically, he asserts that “[o]bstruction of justice ‘is a specific intent crime requiring intent to impair the proceeding,’” relying on Hawkins v. United States, 119 A.3d 687, 695 (D.C. 2015), 10 and that the government did not prove that he intended to impair the proceedings of his contempt trial. We discern no insufficiency on this basis. To begin, we agree with the government’s statement that “obstruction under [this] subsection[] can include retaliation for past events, and does not require intent to impair ongoing, or future, official proceedings.” Cf. Mayhand v. United States, 127 A.3d 1198, 1204 (D.C. 2015) (acknowledging a retaliatory act against an individual “on account of” information given to law enforcement constitutes obstruction under analogous subsection § 22-722(a)(4)); McCullough v. United States, 827 A.2d 48, 58 (D.C. 2003) (holding § 22-722(a)(4) “was satisfied because Hawkins predates this court’s en banc “endorsement of more particularized 10 and standardized categorizations of mens rea” in lieu of “specific” and “general” intent in Carrell, 165 A.3d at 324. 13 [a witness] was killed in retaliation for giving information to the police about criminal activity”). Nonetheless, as Mr. Moore points out, the history of the statute strongly suggests that all obstruction of justice crimes require proof of mens rea of some kind. See The “Law Enforcement Witness Protection Amendment Act of 1992,” D.C. Council, Comm. on the Judiciary, Report on Bill 9-385 at 3 (May 20, 1992) (“Specific knowledge and intent are required for threatening or intimidating conduct to be actionable. The intent requirement embraces in comprehensive terms various forms of obstruction of justice . . . .”). Thus, the fact that obstruction may be accomplished by a retaliatory act does not obviate proof of “specific intent” that we have said is required for this crime; it simply means that an act of retaliation is subject to its own intent requirement, i.e., the intent to retaliate. It does not follow, however, as Mr. Moore argues, that “[t]here can be no obstruction without a threat specifically intended to reach” the AAG who prosecuted him for contempt. Again, assuming in the absence of any persuasive argument to the contrary, see supra section I.B.1, that the government proved the elements of threats against a public official, a rational juror could have concluded that Mr. Moore’s purpose in uttering that threat was to retaliate against the AAG. 14 We thus reject Mr. Moore’s sufficiency arguments about mens rea, and turn to his argument that the government never should have been permitted to present Mr. Harvey’s testimony to the jury to begin with.