In re Andrew J. Kline

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. 13-BG-851

IN RE ANDREW J. KLINE, RESPONDENT.

A Member of the Bar
of the District of Columbia Court of Appeals
(Bar Registration No. 441845)

On Report and Recommendation
of the Board on Professional Responsibility
(BDN 522-09)

(Argued May 22, 2014 Decided April 9, 2015)

Seth A. Rosenthal for respondent.

Elizabeth A. Herman, Deputy Bar Counsel, with whom Wallace E. Shipp,
Jr., Bar Counsel, Jennifer P. Lyman, Senior Assistant Bar Counsel, and Jelani C.
Lowery, Senior Staff Attorney, were on the brief, for petitioner.

Elizabeth J. Branda, Executive Attorney, Board on Professional
Responsibility, for petitioner.

James Klein and Samia Fam, Public Defender Service, filed a brief as
amicus curiae.

Ronald C. Machen Jr., United States Attorney at the time the brief was filed,
Elizabeth Trosman and Ann K. H. Simon, Assistant United Stated Attorneys, Jerri
U. Dunston, Director, United States Department of Justice, and Ann C. Brickley,
Attorney Advisor, Professional Responsibility Advisory Office, United States
Department of Justice, filed a brief as amicus curiae.
 2

Before WASHINGTON, Chief Judge, and GLICKMAN and THOMPSON,
Associate Judges.

WASHINGTON, Chief Judge: This matter comes before us upon the Report

and Recommendation of the Board on Professional Responsibility (“the Board”).

The Board recommended that a 30-day suspension be given to Andrew J. Kline

(“Kline”) after finding that Kline violated Rule 3.8 (e) of the District of Columbia

Rules of Professional Conduct (“Rule 3.8 (e)”). Rule 3.8 (e) prohibits a prosecutor

in a criminal case from intentionally failing to disclose to the defense any evidence

or information that the prosecutor knows or reasonably should know tends to

negate the guilt of the accused. Bar Counsel takes no exception to the Report and

Recommendation of the Board. Kline argued, inter alia, that he did not violate

Rule 3.8 (e) because his ethical duties are coextensive with the duties imposed

under Brady v. Maryland, 373 U.S. 83 (1963). Specifically, Kline relies on the

“material-to-outcome” standard recognized by the United States Supreme Court in

Brady’s progeny to argue that a prosecutor cannot violate Rule 3.8 (e) unless there

is a reasonable probability that the information or evidence withheld made a

difference in the outcome of the trial. We hold that Kline’s interpretation of Rule

3.8 (e), which incorporates a retrospective materiality analysis, is not the

appropriate test for determining whether a prosecutor has violated Rule 3.8 (e).

We also hold that Bar Counsel proved by clear and convincing evidence that Kline

intentionally failed to disclose information in violation of the rule. However, we
 3

conclude that given the confusion regarding the correct interpretation of a

prosecutor’s obligations under the rule, sanctioning Kline would be unwarranted.

I.

The Board adopted the following findings of fact: In 2001 and 2002, Kline

was an Assistant United States Attorney (“AUSA”) in Washington, D.C., assigned

to prosecute violent crimes, including a shooting incident involving Arnell Shelton

(“Shelton”). Shelton was charged with felony assault in the drive-by shooting of

Christopher Boyd (“Boyd”). Prior to trial, Shelton’s attorney filed an alibi notice.

Thus, a principal issue at trial was the reliability of the government’s eyewitnesses.

In the course of preparing the Shelton case, Kline spoke to D.C.

Metropolitan Police Department (“MPD”) Officer Edward Woodward (“Officer

Woodward”), the first officer at the crime scene. The Hearing Committee found

that Officer Woodward told Kline that he first interviewed Boyd at Greater

Southeast Hospital shortly after the shooting, and that Boyd told him that he did

not know who had shot him. Kline took notes during his conversation with Officer

Woodward, and those notes included the following information:
 4

Boyd told officer at hospital that he did not know who
shot him—appeared maybe to not want to cooperate at
the time. He was in pain and this officer had arrested
him for possession of a machine gun. . . .

There is an arrow pointing to this note.1

Shelton’s attorney, Carlos Venegas of the District’s Public Defender Service

(“PDS”), requested discovery pursuant to Brady and specifically sought “prior

inconsistent [or] non-corroborative” statements by witnesses and “any other

information, which . . . impeaches a witness’ testimony.” Kline answered the

Brady request by informing Mr. Venegas that the government was not “in

possession of any truly exculpatory information.”

PDS attorney Anna Rodriques subsequently assumed representation of

Shelton. She testified that Kline never told her of the Boyd Hospital Statement.

Her testimony was corroborated by documentary evidence, including Kline’s

supplemental discovery responses that disclosed other potentially exculpatory

evidence.

1
The Hearing Committee was unable to make a finding as to whether the
arrow is in Kline’s handwriting since Kline did not acknowledge that annotation.
 5

Right before the jury was selected, Attorney Rodriques raised a separate

Brady concern. Kline responded that he was “not sure how one could conjure up a

Brady argument in this case since there was no doubt that Shelton was the

shooter.” The trial court responded:

Because you are sure [sic] you have the guy, no one
could conjure up a Brady argument? . . . That is why
Brady doesn’t leave it up to the prosecutor, for that very
reason. You are always sure you have got the right guy
or you wouldn’t be prosecuting.

Kline assured the trial court that he was “especially careful when it came to Brady

evidence.” However, Kline still had not disclosed the substance of the Boyd

Hospital Statement either directly or indirectly.

The first trial was held between March 5 and 7, 2002. The government’s

case hinged on the ability of the three eyewitnesses to the shooting, Andrew

Durham, Christopher Boyd, and Boyd’s mother, Cassandra Williams, to credibly

identify Shelton as the assailant. Shelton’s wife testified, as an alibi witness, that

he was home at the time of the shooting. The jury was unable to reach a verdict,

and a mistrial was declared.
 6

Soon thereafter, Kline left the United States Attorney’s Office (“USAO”)

and the new AUSA assigned to prosecute the case forwarded the note pertaining to

the Boyd Hospital Statement in Kline’s file. A letter to the defense was prepared

disclosing the information but before it could be mailed, that attorney left the

office due to a family emergency and did not return to the case. When the case

was subsequently reassigned, the new prosecutor, AUSA Wanda Dixon, disclosed

the Boyd Hospital Statement to the defense. A new trial was held and despite the

disclosure of the Boyd Hospital Statement, the defendant was convicted and his

conviction was upheld on appeal.

Because Kline failed to disclose the Boyd Hospital Statement, Bar Counsel

charged him with violating Rule 3.8 (e). Kline, while hedging on whether he in

fact remembered that this evidence was in his file, stated that he did not believe he

had an obligation to turn it over because he did not believe it was Brady evidence.

He also argued that the gist of the statement had been included in police reports

that had been turned over to the defense. He stated further that he believed his

disclosure obligation was only to turn over evidence that fell within the purview of

Brady—i.e. evidence that would prove to be material to the outcome of the trial.

Additionally, Kline presented the testimony of an AUSA responsible for training,
 7

who testified that he was in charge of “Brady” training at the USAO at the time

and while disclosure would have been prudent, the training Kline received from the

U.S. Attorney’s Office on its Brady obligations would not have put [Kline] on

notice that Rule 3.8 (e) required him to disclose information that was not

“material” in the Brady sense.

The Board concluded that Kline violated Rule 3.8 (e) by intentionally

withholding the Boyd Hospital Statement and recommended a sanction of 30 days

suspension. Kline timely appealed.

II.

“The discipline of attorneys, including determination of appropriate

sanctions, is the responsibility of this court.” In re Howes, 39 A.3d 1, 12-13 as

amended nunc pro tunc, 52 A.3d 1 (D.C. 2012) (citation omitted). “Though we

review de novo the Board’s legal conclusions, we must accept the Board’s

evidentiary findings if they are supported by substantial evidence in the record.”

Id. (citation and footnote omitted).
 8

The question of whether and, if so, when a prosecutor’s ethical and

constitutional duties to disclose potentially exculpatory information to a defendant

intersect continues to be a topic of much debate throughout the country. It is

unquestionable, however, that constitutional protections in the criminal context

serve a fundamentally different purpose than disciplinary proceedings in the ethical

context. See, e.g., United States v. Agurs, 427 U.S. 97, 110 (1976) (noting a

distinction between the character of the evidence and the character of the

prosecutor). For the first time, this court must address whether the ethical

disclosure obligations imposed on prosecutors by Rule 3.8 (e) require disclosure of

information that may later be deemed “immaterial” to the outcome of the trial.

Kline argues that Rule 3.8 (e) of the District of Columbia Rules of

Professional Conduct, governing the ethical obligations of prosecutors to disclose

evidence tending to negate the guilt or mitigate the offense of the accused, must be

read as co-extensive with a defendant’s constitutional right to a fair trial as

contemplated by the Supreme Court in Brady, and its progeny. Thus, Kline argues

that the rule necessarily contains a “materiality” component, which cannot be

determined until after trial has been concluded, and the merits of any appeal have

been resolved. More specifically, Kline argues that there can be no violation of
 9

Rule 3.8 (e) unless and until it has been determined that the failure of the

government to disclose any potentially exculpatory information has sufficiently

impacted the fairness of the trial to a degree sufficient to constitute a Brady

violation. This level of unfair prejudice is commonly understood as that which is

“material” to the outcome of a trial.

Kline’s argument regarding “materiality” focuses on the following

definition: whether or not the outcome of the proceeding would have been

different had the evidence or information been disclosed. It is important to note,

however, that this “material-to-outcome” standard was not promulgated in the

landmark Supreme Court case of Brady v. Maryland, but was first formally

adopted years later in United States v. Bagley, 473 U.S. 667 (1985). But see Agurs,

427 U.S. at 107-10 (adopting an approach very similar to the material-to-outcome

test formally adopted in Bagley). This distinction is important because evidence

material to the preparation of the defense is often confused with “material to the

outcome of the trial,” i.e., prejudice. While the Supreme Court in Brady

promulgated a definition of exculpatory material for disclosure purposes—

evidence that is “material to guilt or innocence”—it was not until Bagley that the

term “material” was defined as prejudice sufficient to support a belief that had the

information been disclosed, the outcome of the trial likely would have been
 10

different. See id. at 674-75. This was because, as a reviewing court, the Supreme

Court recognized that reversal of a conviction is unwarranted unless it is

reasonably probable that the evidence withheld would have made a difference at

the trial. See id.

Since the Agurs and Bagley decisions, commentators and legal professionals

often lump the “material-to-outcome” test with Brady because a “true” Brady

violation includes the following three requirements: (1) the evidence at issue must

be favorable to the accused, either because it is exculpatory, or because it is

impeaching; (2) that evidence must have been suppressed by the State, either

willfully or inadvertently; and (3) prejudice must have ensued. See Strickler v.

Greene, 527 U.S. 263, 281-82 (1999).

The ethical rule regarding prosecutorial disclosure in the District of

Columbia, as in most states, incorporated the “tends to negate guilt” standard

promulgated by the ABA in its Model Code of Professional Responsibility to

define the class of evidence required to be disclosed under Rule 3.8. However, the

District of Columbia may be the only jurisdiction in the country that adopted an
 11

“intentionality” requirement as a part of its black letter rule.2 It appears that we

adopted that language from the 1986 ABA Standards for Criminal Justice.

Specifically, Rule 3.8 (e) reads as follows:

The prosecutor in a criminal case shall not: . . . . (e)
Intentionally fail to disclose to the defense, upon request
and at a time when use by the defense is reasonably
feasible, any evidence or information that the prosecutor
knows or reasonably should know tends to negate the
guilt of the accused or to mitigate the offense, or in
connection with sentencing, intentionally fail to disclose
to the defense upon request any unprivileged mitigating
information known to the prosecutor and not reasonably
available to the defense, except when the prosecutor is
relieved of this responsibility by a protective order of the
tribunal.

D.C. R. OF PROF’L CONDUCT R. 3.8 (e) (2014).

The 1986 ABA Standards for Criminal Justice, which is also mentioned with

approval in our commentary to Rule 3.8, provides some guidance on the role Brady

played in the development of the standards for determining what material must be

2
However, at least one state has decided to read an “intentionality”
requirement into its rule. See In re Attorney C, 47 P.3d 1167, 1174 (Colo. 2002).
Alabama has chosen to adopt a “willfulness” standard. See ALA. R. OF PROF’L
CONDUCT R. 3.8 (d) (2014). The ABA rule has no “intentionality” or “willfulness”
requirements by its terms. See MODEL RULES OF PROF’L CONDUCT R. 3.8 (d)
(2013). Rule 3.8 (d) is the ABA counterpart to District of Columbia Rule 3.8 (e).
 12

disclosed to the defendant. Specifically, the commentary notes that, “[t]he

standard adopts the definition of exculpatory material contained in the Supreme

Court’s decision in Brady v. Maryland, that is, material that tends to negate guilt or

reduce punishment. Although the test necessarily presents some questions of

relevance, prosecutors are urged to disclose all material that is even possibly

exculpatory as a prophylactic against reversible error and possible professional

misconduct.” ABA STANDARDS FOR CRIMINAL JUSTICE: The Prosecution Function

§ 3-3.11 (2d ed. 1986). This “tends to negate guilt or mitigate the offense”

standard had made its first prominent appearance in the ethical realm in 1969, in

the ABA Model Code of Professional Responsibility DR 7-103 (b) (“DR 7-103

(b)”), which was promulgated before the Brady prejudice component was defined

in Bagley. See MODEL CODE OF PROF’L RESPONSIBILITY DR 7-103 (b) (1969)

(hereinafter “DR 7-103 (b)”). It stands to reason, therefore, that the later added

prejudice component from Bagley, did not play and could not have played a role in

the development of the “tends to negate guilt or mitigate the offense” standard

promulgated pre-Bagley.

Further, as the Supreme Court recognized in Kyles, “[t]he rule in Bagley

(and, hence, in Brady) requires less of the prosecution than the ABA Standards for

Criminal Justice, which call generally for prosecutorial disclosures of any evidence
 13

tending to exculpate or mitigate.” Kyles v. Whitley, 514 U.S. 419, 437 (1995). The

Supreme Court reiterated that basic tenet in Cone, noting that “[a]lthough the Due

Process Clause of the Fourteenth Amendment, as interpreted by Brady, only

mandates the disclosure of material evidence, the obligation to disclose evidence

favorable to the defense may arise more broadly under a prosecutor’s ethical or

statutory obligations.” Cone v. Bell, 556 U.S. 449, 470 n.15 (2009) (citations

omitted).

Retrospective analysis, while it necessarily comports with appellate review,

is wholly inapplicable in pretrial prospective determinations. See Lewis v. United

States, 408 A.2d 303, 306-07 (D.C. 1979). Specifically, in Lewis, this court

recognized that Brady and its progeny were retrospective evaluations that were

difficult to apply in a pretrial context. “While it is therefore true that the

constitutional question commonly comes up retrospectively, the due process

underpinning of Brady-Agurs is a command for disclosure [b]efore an accused has

to defend himself.” Lewis, 408 A.2d at 306-07. It is impossible for a trial court at

the pretrial stage to require “the defendant . . . to satisfy the test of materiality

normally associated with a retrospective Brady-Agurs inquiry, namely, materiality

to outcome.” See id. at 307. “On the premise that there can be a pretrial ruling

under Brady, this abandonment of the material-to-outcome test is necessary
 14

because there can be no objective, ad hoc way to evaluate before trial whether

[evidence or information] will be material to the outcome. No one has that gift of

prophecy.” Id. Therefore, “[t]o argue that the court can apply a material-to-

outcome test before trial is to argue a contradiction.” Id. (citing Agurs, 427 U.S. at

107-08).

In short, although significant overlaps exist in a pretrial versus post-trial

ethical analysis, it makes little common sense to premise a violation of an ethical

rule on the effect compliance with that rule may have on the outcome of the

underlying trial, because there can be “no objective, ad hoc way” for a prosecutor

“to evaluate before trial whether [evidence or information] will be material to the

outcome.” See Lewis, 408 A.2d at 307. For that reason, it is important not to use

Brady as a “canon of prosecutorial ethics.” Commonwealth v. Tuma, 740 S.E.2d

14, 20 n.2 (Va. 2013).

Kline argues, however, that the last sentence in the comment to Rule 3.8

makes it clear that the “material-to-outcome” test that is ingrained in federal

constitutional law sets forth a prosecutor’s ethical boundaries. The comment to

Rule 3.8 states the following:

Many jurisdictions have adopted the ABA Standards of
Criminal Justice Relating to Prosecution Function, which
 15

in turn are the product of prolonged and careful
deliberation by lawyers experienced in both criminal
prosecution and defense. This rule is intended to be a
distillation of some, but not all, of the professional
obligations imposed on prosecutors by applicable law.
The rule, however, is not intended either to restrict or to
expand the obligations of prosecutors derived from the
United States Constitution, federal or District of
Columbia statutes, and court rules of procedure.

D.C. R. OF PROF’L CONDUCT R. 3.8 cmt. 1 (2014). Kline argues that requiring any

more or less from a prosecutor than is required to avoid a true Brady violation

would alter, expand, or restrict that prosecutor’s obligations under the constitution.

However, Kline’s reliance on the comment to support his interpretation of the rule

is unavailing because the text of the rule is always controlling when it comes to

interpreting a rule. See D.C. R. OF PROF’L CONDUCT Scope (6) (2014) (“[t]he

Comments are intended as guides to interpretation, but the text of each Rule is

controlling.”). Kline’s reading of the commentary—interpreting it as establishing

a Brady materiality test for disclosure—may indeed be what was intended by some

of those who championed inclusion of the limiting language in the comment.3

3
It is regrettable that we lack a contemporaneous explanation for our
court’s decision to add the two sentences to the end of Comment [1]. However,
Kline’s reliance on a single opinion letter as legislative history is unavailing. We
preliminarily note that classifying a letter from the Department of Justice (DOJ)
expressing the opinion of the Office of the Deputy Attorney General written to the
rules committee as “legislative history” sheds no light on the thinking of the
(continued . . .)
 16

However, to the extent Rule 3.8 addresses matters (such as disclosure) that also

are subject to requirements “derived from the United States Constitution, federal or

District of Columbia statutes, [or] court rules of procedure,” the rule should not be

read as an interpretation of those requirements that either expands or contracts

what our court has heretofore (or hereafter) interpreted them to mean. Rather, the

rule governs professional conduct; it may overlap with what constitutional due

process requires, but its requirements are not co-extensive with due process (or

with statutory obligations or court procedural rules).

Moreover, Kline’s argument that the comment imposes a “material-to-

outcome” test on Rule 3.8 (e) not only is inconsistent with the (pre-Bagley) history

(. . . continued)
decision makers and, therefore, must be viewed with some skepticism. Legislative
documents are generated by the “legislature” (or rulemaking body). See BLACK’S
LAW DICTIONARY 919 (8th ed. 1999) (legislative means “of or relating to
lawmaking or to the power to enact laws”). Obviously, this court did not give the
DOJ all that it asked for. Against the DOJ’s wishes, for example, the court kept
the language of paragraph (e) largely intact and refrained from adding an explicit
materiality requirement. Cf. Rule 3.8 (g) (containing explicit materiality
requirement). Moreover, evidence which is of little probative value should not
control over the customary meaning of the words used in the rule and the
comment. See Peoples Drug Stores, Inc. v. District of Columbia, 470 A.2d 751,
755 (D.C. 1983) (“[W]here legislative materials are without probative value, or
contradictory, or ambiguous, they should not be permitted to control the customary
meaning of words.”) (internal quotation marks and alteration omitted).
 17

of the rule but also is counterintuitive when it comes to the development and

implementation of rules designed to guide ethical behavior. “In Brady cases [] an

appellate court sits not as a disciplinary committee of the state bar—but rather as a

court of review, ensuring only that the criminal conviction satisfies the threshold

requirements of due process.” Tuma, 740 S.E.2d at 20 n.2. By contrast, ethical

rules are designed to guide behavior, whereas appellate review of criminal cases is

to ensure, after the fact, that a criminal defendant has received a fair trial. Thus, to

the extent the Rule 3.8 commentary suggests a materiality test, we reject it. We

see no logical reason to base our interpretation about the scope of a prosecutor’s

ethical duties on an ad hoc, after the fact, case by case review of particular criminal

convictions. 4

Only a few state courts have been forced to grapple with this specific issue.

However, a review of the case law from those states is instructive to our analysis.

One case that is factually quite similar to the case at bar, and therefore is

particularly persuasive, is In re Jordan, 913 So. 2d 775 (La. 2005). That case

4
We note that in Virginia, the Virginia Bar rewrote their disclosure rule for
prosecutors, “clarifying that the prosecutor’s ethical duty under that rule is not
coextensive with the prosecutor’s legal duty under Brady.” See Virginia Legal
Ethics Comm. Op. 1862 (2012) (discussing “Timely Disclosure of Exculpatory
Evidence and Duties to Disclose Information in Plea Negotiations”).
 18

involved an eyewitness identification to a murder that occurred at night. Id. at

777-78. The witness stated that she did not have her glasses on the night in

question so she was “coming at this at a disadvantage because she was nearsighted

and needed contacts or glasses for nighttime driving.” Id. at 777. The prosecutor

failed to disclose the witness’s admission that she was nearsighted and was not

wearing her glasses on the night of the murder because the prosecutor “unilaterally

determined that the absence of contacts or glasses on the night of the murder did

not affect” the witness’s identification of the defendant. Id. at 778. During his

disciplinary hearing, the prosecutor testified that he did not violate his ethical

obligations by failing to disclose the information to the defense because he did not

believe the witness’s statement about needing her glasses or contacts was material

and thus, the evidence did not qualify as the type of evidence required to be

disclosed under Brady. Id. at 782.

The Supreme Court of Louisiana concluded otherwise and noted that the

language of Rule 3.8 (d)5 is actually not unlike the “prosecutor’s duty as set forth

in Brady.” Id. at 781. The Supreme Court concluded that because impeachment

evidence is exculpatory (citing Bagley), and because Brady and its progeny require

5
Louisiana’s Rule 3.8 (d) is the analog to our Rule 3.8 (e).
 19

disclosure of exculpatory evidence, the prosecutor had violated Rule 3.8 (d). The

Court went on to specifically reject the notion that the definition of materiality set

forth in Kyles and its progeny can be read as leaving a prosecutor with a degree of

discretion about whether to disclose exculpatory evidence. In concluding that the

material-to-outcome standard in Kyles should not be considered in determining

whether the ethical rule was violated, the court relied in part on Justice Souter’s

statement in Kyles that a prosecutor anxious about “tacking too close to the wind

will disclose a favorable piece of evidence” and “will resolve doubtful questions in

favor of disclosure.” Kyles, 514 U.S. at 439.

In 2012, the Supreme Court of North Dakota reached a similar conclusion,

holding that a prosecutor’s ethical obligation to disclose evidence to the defense is

broader than the duty under Brady or the criminal discovery rule. See In re

Disciplinary Action Against Feland, 820 N.W.2d 672, 678 (N.D. 2012). In

reaching its conclusion that there is a distinction between compliance with an

ethical rule and ensuring that an accused is not wrongly convicted, the Court

looked to the different purposes and objectives served by the two proceedings.

“The primary concern in disciplinary proceedings is to ensure attorneys act in

conformity with the ethical standards embodied in the Rules of Professional

Conduct, regardless of the surrounding circumstances.” Id. “A prosecutor’s
 20

ethical duty to disclose all exculpatory evidence to the defense does not vary

depending upon the strength of the other evidence in the case.” Id. It stands to

reason that “a prosecutor’s failure to comply with the duties imposed by Rule 3.8[]

should not be excused merely because, based upon the other evidence presented at

trial, the result in the case would have been the same.” Id.

While the Supreme Courts of Louisiana and North Dakota have interpreted

the disclosure requirements of prosecutors more broadly, there are courts that have

decided that it would be confusing to prosecutors if they were required to comply

with two different disclosure standards. See In re Riek, 834 N.W.2d 384 (Wis.

2013). This is much the same argument raised by Kline here and was clearly part

of what motivated some members of the committee that developed this ethical rule

to add the Comment to Rule 3.8 (e). Interestingly, the Wisconsin Supreme Court

relied, in part, on the Comment to our rule as support for its decision to limit the

disclosure obligations of prosecutors under their ethical rules to only that

information that later proves to be material to the outcome of the trial. See In re

Riek, 834 N.W.2d at 696. The court also relied on decisions by courts in Ohio,

Louisiana, and Colorado to support its conclusion that the ABA’s Opinion on the

Ethical Duties of Prosecutors that requires the disclosure of all potentially

exculpatory information had not been universally adopted. While the case cited
 21

from Louisiana for this proposition is In re Jordan, and while we disagree that a

fair reading of that case supports the Wisconsin court’s decision, we do

acknowledge that both Colorado and Ohio have reached conclusions consistent

with that of Wisconsin. See In re Attorney C, 47 P.3d 1167 (Colo. 2002) and

Disciplinary Counsel v. Kellogg-Martin, 923 N.E.2d 125 (Ohio 2010).

Contrary to the concerns expressed by those courts, however, we do not

believe that interpreting Rule 3.8 to require greater disclosure than that which may

result in an unfair trial for a criminal defendant will give rise to any confusion

among local prosecutors as to what they are obligated to disclose. First, and

foremost, we find it instructive that all of the prosecutors who later became aware

of the existence of the Boyd Hospital Statement after Kline left the U.S. Attorney’s

Office recognized that the statement was potentially exculpatory and had to be

disclosed. Even Kline’s supervisor, while acknowledging that no specific training

was provided regarding a prosecutor’s obligations under Rule 3.8, testified that it

would have been prudent for Kline to have disclosed the statement. Therefore, and

despite the language in the Comment to Rule 3.8, prosecutors in the USAO were

aware that their disclosure obligations are not determined by whether this court

ultimately concludes that the non-disclosure resulted in a Brady violation.
 22

Further, adopting an ethical rule that errs in favor of disclosure will better

ensure that criminal defendants in the District of Columbia receive a fair trial. All

too often we are asked to decide whether information withheld by the government

was exculpatory and whether that information undermined the fairness of the

criminal trial in that case. Often, the call is a close one, with the court making the

best judgments it can about the impact the exculpatory evidence would have had

on a jury’s verdict or whether the information would have led to other potentially

exculpatory information that might have impacted the jury’s verdict. See, e.g.,

Agurs, 427 U.S. at 108 (describing the materiality test as “an inevitably imprecise

standard”); Boyd v. United States, 908 A.2d 39, 56 (D.C. 2006) (characterizing

evidence in a Brady inquiry as “difficult to assess”); Ginyard v. United States, 816

A.2d 21, 32 (D.C. 2003) (acknowledging as “imprecise” the inquiry into whether

evidence would help the defense or affect the outcome of the trial). These are

judgment calls that can undermine the public’s trust and confidence in the courts

because they are not being made by a jury of one’s peers but by a court that is

sitting and reviewing a cold record. And, even where an appeals court ultimately

decides that the failure of a prosecutor to disclose certain potentially exculpatory

information should result in a new trial, the defendant has already spent a

significant amount of time in jail with the concomitant consequences that

incarceration has on the defendant’s life and that of his or her family. In fact, even
 23

in those instances where the trial court becomes aware of the potentially

exculpatory information or evidence and orders its disclosure during trial, the

adverse impacts on a defendant can be great. See United States v. Stevens, 08-CR-

231 EGS, 2009 WL 6525926 (D.D.C. Apr. 7, 2009) (vacating jury’s guilty verdict

against Senator Ted Stevens on corruption charges where prosecution failed to

produce exculpatory evidence until nearly five months after the trial and after

Senator Stevens narrowly lost his reelection bid).

In Zanders v. United States, 999 A.2d 149 (D.C. 2010), we confirmed that

“[i]t is not for the prosecutor to decide not to disclose information that is on its face

exculpatory based on an assessment of how that evidence might be explained away

or discredited at trial, or ultimately rejected by the fact-finder.” Id. at 164. And

subsequently in Miller v. United States, 14 A.3d 1094 (D.C. 2011), we recognized

that the duty of disclosure is not dependent on whether a defendant’s constitutional

rights are later found to have been violated because the failure to disclose the

information affected the outcome of the trial. Id. at 1109.

While it has not been argued, we are also mindful that were we to adopt

Kline’s interpretation of the comment to reading of Rule 3.8 (e) in a manner

consistent with a prosecutor’s obligations under Brady, et al., the result could have
 24

significant potential adverse impacts for prosecutors generally. For example, in

order to violate Rule 3.8 (e), there must be evidence presented that a prosecutor

intentionally failed to disclose exculpatory evidence. However, a Brady violation

can be “inadvertent.” See Strickler, 527 U.S. at 281-82. Second, Rule 3.8 (e) only

requires disclosure of evidence about which the prosecutor has actual knowledge,

while under Brady potentially exculpatory evidence known by other government

actors is imputed to the prosecution. Third, a violation of Rule 3.8 (e) requires a

finding that the prosecutor knew or reasonably should have known that the

evidence tended to negate the guilt of the accused or mitigate the offense, whereas

a Brady violation is not focused on the conduct of the prosecutor, only whether the

evidence was potentially exculpatory and whether the outcome of the trial was

seriously affected. In sum, Rule 3.8 (e), by its very terms, cannot be read as being

coextensive with Brady and we doubt seriously whether local prosecutors would

support such an interpretation of the rule.

For all of these reasons, we hold that Rule 3.8 (e) requires a prosecutor to

disclose all potentially exculpatory information in his or her possession regardless

of whether that information would meet the materiality requirements of Bagley,

Kyles, and their progeny.
 25

III.

Having determined that a prosecutor’s ethical obligations are not governed

by whether the courts ultimately conclude that a criminal conviction was obtained

in violation of a defendant’s constitutional rights, we turn now to the merits of the

underlying case against Kline. Kline argues that he did not intentionally fail to

disclose the Boyd Hospital Statement, but admits that he cannot say “what [his]

thinking was” nine years ago when he was prosecuting the case against Shelton.

The “standard of clear and convincing proof requires evidence that will produce in

the mind of the trier of fact a firm belief or conviction as to the facts sought to be

established.” In re Dortch, 860 A.2d 346, 358 (D.C. 2004) (citation omitted).

Direct proof of a lawyer’s state of mind is “rarely available.” In re Starnes, 829

A.2d 488, 500 (D.C. 2003) (per curiam).

In the context of other ethical rules, we have adopted various definitions of

“intentional.” See, e.g., In re Mitrano, 952 A.2d 901, 925 (D.C. 2008) (standard

for intentional misappropriation requires a showing that attorney handled the

entrusted funds “in a way that reveals [] an intent to treat the funds as . . . [his]

own”) (citation omitted); In re Ukwu, 926 A.2d 1106, 1116 (D.C. 2007)

(intentional neglect of client’s case “does not require proof of intent in the usual
 26

sense of the word. Rather, neglect ripens into an intentional violation when the

lawyer is aware of his neglect of the client matter”) (internal quotation marks and

citation omitted); In re Lenoir, 585 A.2d 771, 778 (D.C. 1991) (intentional failure

to carry out a contract of employment requires an element of purposefulness or

deliberateness or, at a minimum, an aggravated neglect); In re Reback, 487 A.2d

235, 240 (D.C. 1985), aff’d in relevant part, 513 A.2d 226 (D.C. 1986) (en banc)

(intentional failure to seek a client’s objectives requires an element of

purposefulness or deliberateness or, at a minimum, an aggravated neglect).

We believe that the intentionality requirement under Rule. 3.8 (e) best fits

the definition employed in the context of intentional failures to act—namely, that

“intentional” requires an element of purposefulness or deliberateness or, at a

minimum, of aggravated neglect. See In re Lenoir, 585 A.2d at 778 (citation

omitted). In assessing intent, the “entire mosaic” of conduct should be considered.

In re Ukwu, 926 A.2d at 1117.

The Board argues that there is an “entire mosaic” of circumstances

surrounding the failure to disclose that supports the conclusion that Kline’s failure

to produce the Boyd Hospital Statement was a purposeful or deliberate act. First,

Kline not only spoke to Officer Woodward about the substance of the Boyd
 27

Hospital Statement, but he also wrote the information down on his legal pad

demonstrating that he understood that the victim “told [the] officer at [the] hospital

that he did not know who shot him.”6 Second, Kline also consistently maintained

that he simply did not think the information was exculpatory. (“I can tell you that I

did not at the time think, hmm, this is material evidence that needs to be disclosed

to the defense, nor do I think that as I sit here today. As a matter of fact, I can tell

you based on my experience and my training that it was not and is not material.”).

Third, AUSA Dixon’s statements on the record to the trial court that Kline did not

disclose the information because he did not view the evidence as exculpatory

supports Kline’s own testimony in that regard.7 Fourth, Kline was reminded of his

disclosure obligations on more than one occasion by the trial court during the trial

in this case and was even verbally reprimanded for failing to disclose other

6
The Hearing Committee credited Officer Woodward at the disciplinary
hearing. We find Kline’s challenges to the credibility findings of Officer
Woodward unavailing. See In re Temple, 629 A.2d 1203, 1208-09 (D.C. 1993)
(“The fact-finder who hears the evidence and sees the witnesses is in a better
position to make [credibility] determinations, having the benefit of those critical
first-hand observations of the witness’ demeanor or manner of testifying which are
so important to assessing credibility.”).
7
Although it appears that AUSA Dixon may have gotten that information
from her supervisors and not directly from Kline, this statement though minimally
probative, still adds to the quantum of evidence presented to satisfy the clear and
convincing hurdle. The statement is probative, however, because it corroborates
Kline’s own statements in that regard—namely, that he just did not think this
material was exculpatory.
 28

potentially exculpatory information. Despite the fact that he was on notice that the

trial court was concerned about his failure to disclose other information in the case,

he still did not disclose the Boyd Hospital Statement, instead assuring the trial

court that he was “especially careful” when it came to the disclosure of Brady

information. Further, the Board found that Kline’s testimony about his lack of

knowledge about the Boyd Hospital Statement was less than convincing.

On appeal, Kline argues that he does not remember whether he consciously

thought about the information. However, before the hearing committee he testified

that he knew Boyd did not make an identification of Shelton at the hospital, but

blamed his terrible note taking for misunderstanding the importance of that

testimony and interpreting it to mean that Boyd was merely unable to tell the

officer who shot him at that time for reasons associated with his having been shot.

Kline also testified that because he believed the statement was ambiguous, he did

“not recognize [it] as exculpatory.” He further testified that he believed the

information had been effectively turned over because the police reports disclosed

“97.7%” of the information.

After reviewing the entire record, we see no reason to disturb the findings of

the Hearing Committee and the Board that Kline consciously decided that the
 29

Boyd Hospital Statement did not have to be produced and thus acted with

“deliberateness.” See In re Lenoir, 585 A.2d at 778. Therefore, we agree that the

evidence is such that it produces in the mind of the trier of fact a “firm belief” that

Kline intentionally withheld the statement because he did not think it was

exculpatory. See In re Dortch, 860 A.2d at 358.

Sanction

“In disciplinary cases, the Board must accept the Hearing Committee’s

evidentiary findings, including credibility findings, if they are supported by

substantial evidence in the record.” In re Ukwu, 926 A.2d at 1115. “This court, in

turn, must accept the Board’s findings of fact, and we also apply the ‘substantial

evidence’ standard.” Id.

In In re Howes, this court dealt with a violation of Rule 3.8 (e) for the first

time; however, we did not have to decide whether an ethical violation of Rule 3.8

(e) was dependent on whether the nondisclosure resulted in a Brady violation

because the prosecutor in that case stipulated that he had violated the rule. See In

re Howes, 39 A.3d at 1 n.1. Thus, this is the first opportunity this court has had to

decide the scope of a prosecutor’s ethical responsibilities under Rule 3.8 (e).
 30

“Generally speaking, if the Board’s recommended sanction falls within a

wide range of acceptable outcomes, it will be adopted and imposed. We grant

deference to the recommended disposition of the Board unless to do so would

foster a tendency toward inconsistent dispositions for comparable conduct or

would otherwise be unwarranted. However, the responsibility for imposing

sanctions rests with the court in the first instance.” Id. at 13 (internal citations and

quotations omitted).

While the issue of an appropriate discipline for a prosecutor who violates his

disclosure obligations under Rule 3.8, but who is not found to have been dishonest,

is res nova in the District of Columbia, other jurisdictions have imposed discipline

that range from public reprimand or censure to a six-month suspension from the

practice of law.8 Here, the Board recommends that Kline be suspended for 30

days, a sanction that is clearly within the wide range of sanctions that generally

8
See, e.g., In re Jordan, 91 P.3d 1168 (Kan. 2004) (public censure); In re
Grant, 541 S.E.2d 540 (S.C. 2001) (public reprimand by consent); Committee on
Prof’l Ethics & Conduct of Iowa State Bar Ass’n v. Ramey, 512 N.W.2d 569 (Iowa
1994) (indefinite suspension with no possibility of reinstatement for three months);
Office of Disciplinary Counsel v. Jones, 613 N.E.2d 178 (Ohio 1993) (six-month
suspension).
 31

would be appropriate.9 However, while clear and convincing evidence has been

presented that Kline violated Rule 3.8 when he failed to turn over the Boyd

Hospital Statement to the defense prior to trial, we are mindful of the fact that our

comment to Rule 3.8 (e) has created a great deal of confusion when it comes to a

prosecutor’s disclosure obligations under Rule 3.8. Indeed, the ABA issued a

formal opinion10 on this topic and interpreted our comment to mean that Brady

9
An appropriate sanction is one that is necessary to protect the public and
the courts, maintain the integrity of the profession, and deter other attorneys from
engaging in similar misconduct. See In re Kline, 11 A.3d 261 (D.C. 2011) (citing
In re Reback, 513 A.2d at 231). In determining the appropriate sanction the Court
considers: (1) the nature and seriousness of the misconduct, (2) the presence of
misrepresentation or dishonesty, (3) the respondent’s attitude toward the
underlying misconduct, (4) prior misconduct, (5) mitigating or aggravating
circumstances, and (6) prejudice to the client. See In re Hutchinson, 534 A.2d 919,
924 (D.C. 1987) (en banc).
10
See American Bar Association Formal Opinion, Prosecutor’s Duty to
Disclose Evidence and Information Favorable to the Defense at 4 n.18 (July 8,
2009). The opinion states, in pertinent part:

Rule 3.8 (d) sometimes has been described as codifying the Supreme
Court’s landmark decision in Brady v. Maryland, which held that
criminal defendants have a due process right to receive favorable
information from the prosecution. This inaccurate description may
lead to the incorrect assumption that the rule requires no more from a
prosecutor than compliance with the constitutional and other legal
obligations of disclosure, which frequently are discussed by courts in
litigation. . . . The ABA adopted the rule against the background of
the Supreme Court’s 1963 decision in Brady v. Maryland, but most
understood that the rule did not simply codify existing constitutional
law but imposed a more demanding disclosure obligation.
 32

materiality, in the “material-to-outcome” sense, was required to find an ethical

violation of Rule 3.8 (e). And, as recently as last year, the Supreme Court of

Wisconsin, relying in part on the Comment to our Rule 3.8 as well as the

interpretation given to it by the 2009 ABA Formal Opinion, held that a

prosecutor’s disclosure duties under the ethical rules were co-extensive with their

obligations under Brady and thus, there could be no violation of the ethical rule

unless the court finds that a Brady violation occurred. See In re Riek, 834 N.W.2d

at 390. When we add in the testimony of an AUSA responsible for training that

the U.S. Attorney’s Office did not provide any separate training on a prosecutor’s

Rule 3.8 (e) disclosure obligations, and the argument by Kline that he understood a

prosecutor’s ethical obligations to be coextensive with his obligations under Brady

and that no violation of Rule 3.8 (e) can be found independent of a Brady violation,

we must conclude that his understanding was wrong but it was not unreasonable,

and that no sanction is warranted.

In so concluding, we are also taking into consideration no companion

violations were charged, no allegations of dishonesty were made, the respondent

has a clean disciplinary record, and similar conduct will incur sanctions

comparable to that recommended by the BPR in this case now that this court has
 33

provided clear guidance on the scope of a prosecutor’s disclosure obligations under

Rule 3.8.

So ordered.