United States Court of Appeals
FOR THE DISTRICT OF COLUMBIA CIRCUIT
Argued May 5, 2006 Decided July 7, 2006
No. 04-7013
CHARLES SINGLETARY,
APPELLANT
v.
EDWARD F. REILLY, JR., IN HIS OFFICIAL CAPACITY AS
COMMISSIONER OF THE U.S. PAROLE COMMISSION, ET AL.,
APPELLEES
Appeal from the United States District Court
for the District of Columbia
(No. 00cv01263)
Catharine F. Easterly, Assistant Public Defender, Public
Defender Service of the District of Columbia, argued the cause
for appellant. With her on the brief was Timothy P. O'Toole,
Assistant Public Defender.
Elizabeth H. Danello, Assistant U.S. Attorney, argued the
cause for appellees. With her on the brief were Kenneth L.
Wainstein, U.S. Attorney, and Roy W. McLeese, III and Thomas
J. Tourish, Jr., Assistant U.S. Attorneys.
Before: SENTELLE and BROWN, Circuit Judges, and
EDWARDS, Senior Circuit Judge.
2
Opinion for the Court filed by Circuit Judge BROWN.
BROWN, Circuit Judge: Charles Singletary’s parole was
revoked in 1996 based on his alleged participation in a murder.
The evidence tying Singletary to this offense—for which he has
been imprisoned for the last ten years—consisted solely of
hearsay testimony relayed by a prosecutor and an investigating
detective. The reliability of the hearsay, most of it multilayered,
was never established, and its accuracy remains open to serious
questions. A parole revocation hearing is not a criminal trial, and
the same standards of proof and admissibility of evidence do not
apply. Yet though the government is not required to carry a
heavy burden in such proceedings, it cannot return a parolee to
prison based on a record as shoddy as this one. We therefore
conclude that Singletary is entitled to a new parole revocation
hearing.
I
Singletary was originally sentenced to nine to twenty-seven
years’ imprisonment (minus 388 days) for robbery, armed
robbery, and assault with a deadly weapon. He began serving his
sentence on January 28, 1983, and was granted parole on June
1, 1990. In June 1995, Singletary was arrested for the murder of
Leroy Houtman, a.k.a. Vaughn Stokes, but the case was never
brought before a grand jury, and Singletary was eventually
released without being charged. However, on July 30, 1996, the
District of Columbia Board of Parole held a joint hearing to
consider parole revocation for Singletary and his alleged
accomplice, Gary Barnes, based on the Houtman murder.1
1
For Barnes, this hearing served as a continuation of a previous
hearing held several months earlier.
2
3
We cannot be sure of everything that transpired at the
hearing. Singletary has provided us with a partial transcript
created by his counsel from an audiotape of the proceedings.
However, the transcript cuts off in the middle of a sentence, and
Singletary claims that he has never been provided with a
recording or transcript of the rest of the hearing. Our only other
information regarding the events of that day comes from the
Board’s semi-legible handwritten findings of fact and
Singletary’s abbreviated description of the hearing, neither of
which provides any significant additional details.
Singletary was charged with failing to obey all laws and
using a deadly weapon in connection with first degree murder.
He denied both accusations. The Board began by asking Peter
Zeidenberg, an Assistant United States Attorney, if he “could
tell [the Board] what [he] kn[e]w about this case.” Zeidenberg
presented his view of what “the evidence was in this case”
through his own narrative as well as by questioning Detective
Todd Amis, who investigated Houtman’s murder. Most of the
information described by Zeidenberg and Amis originally came
from two individuals—Verdez Smith and Terri Washing-
ton—who were not identified by name during the hearing,
though their names have been revealed during this litigation.
Smith and Washington claimed to have learned about the
murder from Carmelita Metts, who was convicted of conspiracy
to murder Houtman.
According to Zeidenberg and Amis, Houtman’s body was
discovered in Metts’s apartment complex when plumbers
working in a utility room smelled an odor and saw a foot
sticking out from some plastic. Houtman had been stabbed fifty-
one times; his body had been wrapped in insulation that was
pulled down from the ceiling. He had been dead for approxi-
mately three weeks.
4
Amis stated that the body was finally identified three
months later by Houtman’s sisters, based on a piece of jewelry
Houtman was wearing. Amis then began a search for Houtman’s
missing truck, which bore fruit when he discovered it in Smith’s
possession. Smith initially claimed that someone named “Tony”
gave him the truck; he later admitted to receiving it from Metts,
although he claimed not to know where she obtained it. After
being charged with Houtman’s murder, Smith changed his story
again, claiming that Metts—his former girlfriend—had asked
him to help her kill Houtman, but that he had refused. Smith
claimed that Metts had later given him Houtman’s truck so that
he could sell it, as he had previous experience as a car thief. The
charges against Smith were dropped.
The police also interviewed Washington, who initially
denied knowing anything about the murder (as Amis stated in a
sworn statement made after the hearing). Amis told the Board
that Washington then claimed Metts confessed to helping Barnes
and Singletary murder Houtman.2 Allegedly, Metts lured
Houtman to her apartment, where Barnes and Singletary (and
possibly Metts herself) stabbed him, stored the body in the
closet temporarily, and later moved it to the basement. Amis
informed the Board that Smith also claimed Metts told him of
this sequence of events.3
2
At Metts’s trial, Washington testified that she was dating
Houtman prior to his disappearance; she subsequently met and started
dating Barnes. Thus, in a rather bizarre love triangle, Washington
either dated her ex-boyfriend’s murderer or—as Singletary
implies—helped murder her boyfriend and then framed her new
boyfriend for the crime.
3
Additionally, Zeidenberg stated—apparently recapping Smith’s
testimony at trial—that Smith went to Metts’s apartment after the
murder and noticed that Metts “looked like she hadn’t slept.” Smith
allegedly told his ex-girlfriend that she “look[ed] terrible,” to which
5
Zeidenberg stated that the police searched Metts’s apart-
ment and found some of Houtman’s blood. He stated that when
Metts found the notice left on her door by the police, she was
with Washington, who then accompanied her to the barber shop
where Singletary worked. Zeidenberg and Amis both stated that
Washington claimed to have seen Singletary wearing a distinc-
tive ring that had belonged to Houtman.(However, at Metts’s
trial, Washington actually testified that Singletary was wearing
Houtman’s bracelet, and that she saw Barnes wearing
Houtman’s ring at one point. Neither piece of jewelry was ever
recovered by the police.) According to Zeidenberg, Metts told
Washington and Smith she gave Houtman’s keys to Barnes and
Singletary in order for them to search his apartment for drugs
and money, but that they did not give her a share of whatever
they found.
Singletary was eventually arrested. After his arrest, Amis
asked if he knew Metts or Barnes, but Singletary allegedly
denied knowing either. Amis later found a picture of Singletary
and Barnes together at a wedding. Singletary’s attorney had an
opportunity to cross-examine Amis (although only part of this
exchange is reflected in the hearing transcript). When asked if
the police had “anything, anything at all besides double hearsay
that connects [Singletary and Barnes] to the murder,” Amis
simply stated, “no.” Amis also admitted that no murder weapon
had been found.
Zeidenberg acknowledged Singletary’s case had never been
presented to a grand jury because the prosecution’s only
she replied, “Well, you’d look bad too if you hadn’t slept for two
days.” Metts allegedly said she had “had a dead body in [her] closet
for the last two days,” but showed Smith that it wasn’t there anymore.
She then allegedly showed Smith the padlocked door to the utility
room in the basement where the body had been concealed.
6
admissible evidence would have been Metts’s statements. The
prosecution did not want to “cut a deal” with Metts to get her to
testify; “any confession that she gave would be so suspect that
. . . it would [be] counterproductive.” According to Singletary’s
former attorney, prior to the hearing, the Board did not inform
him of any evidence, witness statements, or other information
regarding the alleged parole violations. He claimed that
Singletary brought two “favorable witnesses” to the hearing, but
the Board did not permit them to enter the hearing room. The
record does not reflect that Singletary complained about either
issue during the hearing.
The Board revoked Singletary’s parole on August 6, 1996.
Singletary filed for a writ of habeas corpus in 1997, but the writ
was denied by the D.C. Superior Court, and the D.C. Court of
Appeals affirmed. Singletary v. Quick, No. 97-SP-1984 (D.C.
July 24, 1998) (unpublished order). He again sought habeas
relief in 2000, but again the Superior Court denied his claims,
and again the Court of Appeals affirmed. Singletary v. D.C. Bd.
of Parole, 794 A.2d 56 (D.C. 2001) (unpublished table deci-
sion). Finally, Singletary petitioned for a writ of habeas corpus
in the U.S. District Court for the District of Columbia, and his
petition was denied one more time. Singletary v. D.C. Bd. of
Parole, No. 00-cv-01263 (D.D.C. Dec. 16, 2003) (unpublished
opinion). The district court found that “the only challenge raised
by petitioner through his counsel at the parole revocation
hearing that is now being raised was the challenge to the
sufficiency of the evidence.” Id. at 7. The court thus considered
other challenges brought by Singletary forfeited, a decision that
Singletary does not challenge on appeal. Finding the hearsay
sufficiently reliable, the district court denied Singletary’s
petition. Id. at 8-10.4
4
On appeal, we granted the Board’s motion to substitute the
United States as its successor in interest, given that the Board no
7
II
In Morrissey v. Brewer, 408 U.S. 471, 487-88 (1972), the
Supreme Court mandated a hearing “prior to the final decision
on revocation by the parole authority,” if the parolee desires
such a hearing. The hearing
must be the basis for more than determining probable cause;
it must lead to a final evaluation of any contested relevant
facts and consideration of whether the facts as determined
warrant revocation. The parolee must have an opportunity
to be heard and to show, if he can, that he did not violate
the conditions, or, if he did, that circumstances in mitigation
suggest that the violation does not warrant revocation.
Id. at 488. While declining to “write a code of procedure” for
such hearings, the Court did specify that “the minimum require-
ments of due process” included:
(a) written notice of the claimed violations of parole; (b)
disclosure to the parolee of evidence against him; (c)
opportunity to be heard in person and to present witnesses
and documentary evidence; (d) the right to confront and
cross-examine adverse witnesses (unless the hearing officer
specifically finds good cause for not allowing confronta-
tion); (e) a “neutral and detached” hearing body such as a
traditional parole board, members of which need not be
judicial officers or lawyers; and (f) a written statement by
the factfinders as to the evidence relied on and reasons for
revoking parole.
longer exists and the United States consented to our exercise of
jurisdiction. Singletary v. D.C. Board of Parole, No. 04-7013 (D.C.
Cir. Apr. 24, 2006) (unpublished order).
8
Id. at 488-89. The hearing process “should be flexible enough to
consider evidence including letters, affidavits, and other material
that would not be admissible in an adversary criminal trial.” Id.
at 489. See also Gagnon v. Scarpelli, 411 U.S. 778, 782 n.5
(1973) (“[W]e did not in Morrissey intend to prohibit use where
appropriate of the conventional substitutes for live testimony,
including affidavits, depositions, and documentary evidence.”).
Parole revocation violates due process if the decision is
“either totally lacking in evidentiary support or . . . so irrational
as to be fundamentally unfair.” Duckett v. Quick, 282 F.3d 844,
847 (D.C. Cir. 2002). In Crawford v. Jackson, 323 F.3d 123, 128
(D.C. Cir. 2003), we found that “[r]eliance on hearsay in parole
revocation hearings is not per se impermissible.” However, “the
use of unsubstantiated or unreliable hearsay would certainly
eviscerate the safeguards guaranteed by Morrissey and
Gagnon.” Id. (internal quotation marks and ellipses omitted).
Rather than focusing on whether evidence would be admissible
at a criminal trial, courts reviewing revocation decisions “are
properly more concerned with whether the evidence considered
as a whole, including the hearsay evidence, was both sufficient
in quantity and reliability to ensure fundamental due process
rights.” Id. We applied the Duckett standard for the sufficiency
of the evidence and “follow[ed] other circuits that have exam-
ined the reliability of the particular hearsay evidence, condemn-
ing reliance on it when the court reaches a negative evaluation.”
Id. at 129.
Other circuits have split on the question of whether, prior to
admitting hearsay, the parole authority must make an explicit
finding of good cause for not allowing a parolee to confront an
adverse witness. See, e.g., Barnes v. Johnson, 184 F.3d 451, 454
(5th Cir. 1999) (“[T]he hearing officer must make an explicit,
specific finding of good cause and state the reasons for that
finding.”); Egerstaffer v. Israel, 726 F.2d 1231, 1234 (7th Cir.
9
1984) (stating that no explicit finding is required when hearsay
evidence “bears substantial guarantees of trustworthiness”).
While we have not required an explicit finding of good cause at
the hearing, we have placed the burden on the “parole authori-
ties to ensure, before relying on hearsay, that there are sufficient
indicia of reliability under the circumstances at hand to protect
the prisoner’s due process rights.” Crawford, 323 F.3d at 129.
We expressed “concern about the reliance in parole revocation
hearings on hearsay in police reports” and warned about the
consequences of relying on unreliable hearsay:
[G]iven judicial concern in light of the protections guaran-
teed by Morrissey, a parole authority takes a certain risk
that its decision to revoke parole will not be judicially
sustained where it relies solely on hearsay contained in a
police investigative report as the basis for its decision. . . .
[T]hat risk is measurably lessened only in circumstances
that demonstrate the strong reliability of the hearsay
evidence.
Id. See also Ash v. Reilly, 431 F.3d 826, 829-30 (D.C. Cir. 2005)
(explaining that our Crawford v. Jackson decision still governs
parole revocation hearings even after Crawford v. Washington,
541 U.S. 36 (2004), which held that use of “testimonial”
hearsay at a criminal trial violates a defendant’s Sixth Amend-
ment rights unless the witness is unavailable and the defendant
had a prior opportunity to cross-examine him).
In our Crawford case, the Board relied on a police investi-
gative report in revoking a prisoner’s parole. We found that the
report was “quite detailed, an indicia of reliability,” as it
provided “a fairly full account of the circumstances surround-
ing” the parole violation. Crawford, 323 F.3d at 130. We also
found several other factors to be relevant: the report was
corroborated on some points by the parolee’s admissions at the
10
hearing and contained “internal corroboration” of the alleged
violation. Id. Next, the parolee was given a chance to present
contrary evidence at the hearing, but he chose not to do so, even
though he was not denied access to the report and knew the
identities of the relevant witnesses. Id. Under those circum-
stances, “given the indicia of reliability of the police investiga-
tive report, the Board’s reliance on the hearsay evidence did not
render its revocation decision so lacking in support that it was
fundamentally unfair.” Id. at 131.
III
While the hearsay evidence at issue in Crawford “was both
sufficient in quantity and reliability to ensure fundamental due
process rights,” id. at 128, a far different case is presented here.
We have no cause to question the quantity of the evidence
presented at the hearing; if its reliability had been established,
the Board’s revocation decision would most likely not be
problematic. Yet the issue we confront today is not the quantity
of the hearsay presented, but rather its reliability. We find that
the hearsay presented at the hearing was not demonstrated to be
reliable and that the Board’s decision to revoke Singletary’s
parole was therefore “totally lacking in evidentiary support.” Id.
at 129 (citation omitted).
Almost all the evidence presented at the hearing was
hearsay, much of it multilayered. Zeidenberg and Amis de-
scribed, in their own words, what Washington and Smith had
previously stated about Houtman’s murder during the police
investigation and Metts’s trial. The most important
evidence—Metts’s alleged confessions to Smith and Washing-
ton—involves a second layer of hearsay, as Zeidenberg and
Amis were only telling the Board what Smith and Washington
claimed Metts told them.
11
The government emphasizes that Smith and Washington
made statements under oath at Metts’s trial, under penalty of
perjury and subject to cross-examination. While this argument
does suggest some degree of reliability, it is an incomplete
rationale. Smith and Washington were only subject to cross-
examination by Metts’s attorney; while Metts may have had the
motivation to question their general credibility, Singletary’s
identity as a co-conspirator was not a central issue. In fact,
Singletary was not mentioned in the brief cross-examination of
either witness at the trial. The record also reflects that Smith and
Washington both changed their stories after initially denying any
knowledge about the murder. Given these somewhat suspicious
circumstances, there is room to doubt that they accurately
identified Singletary, rather than some other individual, as
assisting Metts.5 Yet Zeidenberg and Amis did not reveal
Smith’s and Washington’s identities at the hearing.6 “Thus the
[B]oard had no way of knowing how reliable the informants
were and had no real basis on which to resolve the credibility
issue against the parolee . . . .” Birzon v. King, 469 F.2d 1241,
1244 (2d Cir. 1972) (finding error when a parole board relied on
anonymous statements in a report rather than receiving testi-
mony directly).
In Crawford, we found that a high level of detail in a
hearsay statement can indicate reliability. 323 F.3d at 130. Metts
allegedly did describe some details of the murder to Smith and
Washington (such as what was done with the corpse), but other
5
Indeed, Singletary implies (albeit without overwhelming
support) that Washington and Smith may have been attempting to shift
blame for the murder away from themselves.
6
The record before this court does not indicate that the Board was
informed of Smith’s and Washington’s identities while Singletary was
not present.
12
details are notably absent from the record (such as the date of
the murder). More significantly, many details regarding
Singletary’s involvement were never explained, including why
he would participate in the murder and whether he even knew
Houtman. At the same time, some of the details that Smith and
Washington did provide were not accurately reflected at the
hearing. For example, Washington testified that she saw
Singletary wearing Houtman’s bracelet and Barnes wearing
Houtman’s ring, yet Zeidenberg and Amis both told the Board
that Washington saw Singletary wearing the “distinctive” ring.
The government contends Singletary’s failure to present
contrary evidence at his hearing supports a finding that the
hearsay was reliable. Yet Singletary’s attorney has stated in an
affidavit that Singletary did bring two witnesses, but they were
not allowed to enter the hearing room. He also stated that, until
the hearing, the government gave him no information about the
charges or the evidence to be presented. The government has not
contradicted these assertions, which, if true, would have made
it rather difficult for Singletary to present any rebuttal evidence.
The government argues Singletary failed at the hearing to object
to the alleged exclusion of his witnesses and therefore forfeited
this issue. Even if he did forfeit his ability to argue that the
exclusion of his witnesses violated his due process rights,
however, we surely will not draw the opposite conclusion—that
he deliberately chose not to present evidence—especially given
that the burden is on the government to “ensure . . . that there are
sufficient indicia of reliability” before relying on hearsay. Id. at
129.
Hence, the government has not established that the hearsay
deemed adequate by the Board was “sufficient in . . . reliability
to ensure fundamental due process rights.” Id. at 128. Smith and
Washington, who apparently changed their stories at least once
each, provided the key information relied on by the Board in
13
revoking Singletary’s parole. Yet the two hearsay declarants
were never cross-examined on the issue of Singletary’s
involvement—let alone cross-examined at the hearing—nor
were their identities even revealed for purposes of evaluating
their credibility. The record before us contains notable inconsis-
tencies and omissions, and a less than complete account of the
hearing itself.
We emphasize that we are not deciding whether the
evidence would have been sufficient to merit parole revocation
if its reliability had been established or if it had been presented
in a more trustworthy format. On this record, though, we are
unable to conclude that the Board properly relied on the hearsay
in question. Absent this hearsay evidence, the Board’s decision
was “totally lacking in evidentiary support.” Id. at 129 (quoting
Duckett, 282 F.3d at 847). We therefore reverse the judgment of
the district court and remand with instructions that Singletary be
provided with a new parole revocation hearing consistent with
this opinion.
So ordered.