United States v. Ellis

UNITED STATES, Appellee
v.
Daniel L. ELLIS, Staff Sergeant
U.S. Air Force, Appellant
No. 09-0382
Crim. App. No. 37113
United States Court of Appeals for the Armed Forces
Argued November 9, 2009
Decided February 23, 2010
ERDMANN, J., delivered the opinion of the court, in which
EFFRON, C.J., and STUCKY and RYAN, JJ. joined. BAKER, J., filed
a separate opinion concurring in the result.
Counsel
For Appellant: Major Michael A. Burnat (argued); Major Shannon
A. Bennett (on brief).
For Appellee: Major Coretta E. Gray (argued); Colonel Douglas
P. Cordova, Lieutenant Colonel Jeremy S. Weber, and Gerald R.
Bruce, Esq. (on brief).
Military Judge: Gregory Gaudette
This opinion is subject to revision before final publication.
United States v. Ellis, No. 09-0382/AF
Judge ERDMANN delivered the opinion of the court.
Staff Sergeant Daniel L. Ellis entered guilty pleas to a
number of offenses involving his inappropriate conduct with a
young girl and an individual he believed to be a young girl.1 A
military judge accepted Ellis’s pleas and sentenced him to
eleven years of confinement, forfeiture of all pay and
allowances, reduction to the grade of E-1, and a dishonorable
discharge. The convening authority approved the sentence and
the United States Air Force Court of Criminal Appeals affirmed
the findings and the sentence. United States v. Ellis, No. ACM
37113, 2008 CCA LEXIS 507, 2008 WL 5192458 (A.F. Ct. Crim. App.
Dec. 12, 2008) (unpublished).
“In a sentencing hearing, an accused’s potential for
rehabilitation is a proper subject of testimony by qualified
experts.” United States v. Stinson, 34 M.J. 233, 238 (C.M.A.
1992) (citing Rule for Courts-Martial (R.C.M.) 1001(b)(5)).
Mil.R.Evid. 702-705 and 403 operate to establish
a simple four-part test for admissibility of expert
testimony: (1) Was the witness “qualified to testify
as an expert”? (2) Was the testimony “within the
limits of [the expert’s] expertise”? (3) Was the
“expert opinion based on a sufficient factual basis to
make it relevant”?, and (4) “Does the danger of unfair
prejudice created by the testimony outweigh its
probative value?” United States v. Stinson, 34 M.J.
233, 238 (C.M.A. 1992); United States v. Neeley, 25
1
Ellis was convicted of two specifications of indecent acts on
divers occasions with a child, one specification of possession
of child pornography, one specification of adultery, four
specifications of communicating indecent language to a child,
one specification of carnal knowledge, and two specifications of
attempted communication of indecent language to a child.
2
United States v. Ellis, No. 09-0382/AF
M.J. 105, 107 (C.M.A. 1987), cert. denied, 484 U.S.
1011 (1988).
United States v. Banks, 36 M.J. 150, 161 (C.M.A. 1992).
We granted review in this case to determine whether the
military judge abused his discretion by allowing the
Government’s expert to testify as to Ellis’s potential for
rehabilitation, specifically his risk of recidivism.2 We hold
that the military judge did not abuse his discretion in allowing
the expert opinion testimony and therefore affirm the Air Force
Court of Criminal Appeals.
Background
Ellis was stationed at Elmendorf Air Force Base near
Anchorage, Alaska. In December 2004 he met VC, then a thirteen-
year-old female seventh grader, in an Internet chat room
operated by Yahoo. During the course of their chats, VC gave
Ellis her name and age and told him that she lived in Anchorage.
Ellis continued these chats with VC almost every other day,
which ultimately led to a meeting with VC and her mother at an
Anchorage restaurant. During that meeting, VC’s mother told
Ellis that VC was only thirteen years old.
About two weeks later, Ellis went to VC’s home and again
visited with VC and her mother. VC’s mother left the house and
Ellis and VC participated in what VC described as a “make out
session.” At some point later, Ellis met VC’s mother at her
2
United States v. Ellis, 68 M.J. 145 (C.A.A.F. 2009) (order
granting review).
3
United States v. Ellis, No. 09-0382/AF
house and they engaged in sexual intercourse. During this
period Ellis continued to have online chats with VC about the
sexual things he would like to do to her. He also continued his
visits to her house and during these visits Ellis began to have
intimate sexual contact with VC, which ultimately led to sexual
intercourse. On one occasion Ellis took VC to the parking lot
of a local elementary school, where he digitally penetrated her
vagina while sitting in his car.
In October 2005, Ellis was transferred to Cannon Air Force
Base in New Mexico, but he continued his Internet chats and
sexual banter with VC. The sexual chat banter was eventually
discovered and a search warrant obtained to search his home and
personal computer in New Mexico. An analysis of the computer
revealed a number of files containing child and adult
pornography as well as chat logs between Ellis and VC and chat
logs between Ellis and someone he believed to be a fourteen-
year-old female named “Mandy.” Actually “Mandy” was an
Immigration and Customs Enforcement (ICE) officer conducting an
Internet child pornography investigation who was posing as a
child. The chats with the ICE officer contained indecent sexual
banter and took place one month after the initial charges had
been preferred against Ellis in the instant case.
Ellis entered into a pretrial agreement with the convening
authority in return for his guilty pleas. During the
presentencing phase of Ellis’s court-martial, the prosecution
4
United States v. Ellis, No. 09-0382/AF
called an expert, Dr. Timothy Faye Branaman, to testify as to
Ellis’s risk for recidivism. Following a series of questions
concerning Dr. Branaman’s qualifications and experience, and
after his curriculum vitae was admitted into evidence, the trial
counsel moved for his recognition as an expert in forensic
psychology with a specialization in sexual offender assessments.
The defense did not object to Dr. Branaman’s qualifications or
his recognition as an expert.
Dr. Branaman testified that the methodology he used in
performing risk assessments of sexual offenders was an actuarial
approach (statistical degree of probability) using an instrument
entitled Static 99.3 The Static 99 was developed from
statistical studies of men released from incarceration who had
been convicted of sexual offenses. The instrument grouped the
risk factors of these men and compared those factors with the
men’s recidivism rates. Dr. Branaman testified that the
instrument was found to have a seventy percent rate of
predictive validity and was well accepted within the scientific
and medical communities. Static 99 evaluates ten separate risk
factors and assigns points depending on an individual’s history
3
Static 99 is an instrument that was developed and published in
1999 and is widely utilized in this field. See United States v.
McIlrath, 512 F.3d 421 (7th Cir. 2008), and cases cited therein.
The term “static” refers to the historical factors the
assessment evaluates.
5
United States v. Ellis, No. 09-0382/AF
as to each factor.4 Dr. Branaman testified that there is a
possible total of twelve points on the Static 99: zero and one
reflect a low risk of recidivism; two and three reflect a
moderate low risk; four and five reflect a moderate high risk;
and six and above reflect a high risk.
Prior to rendering his opinion, Dr. Branaman testified that
he had reviewed the charges and specifications against Ellis;
reviewed the stipulation of fact; reviewed the forensic analysis
of items seized from Ellis’s computer and the chat logs;
listened to the guilty plea inquiry by the military judge; and
reviewed the rehabilitation options available at Cannon Air
Force Base with confinement officials. Dr. Branaman did not
conduct a personal interview of Ellis.
When the prosecution solicited Dr. Branaman’s opinion as to
Ellis’s risk of recidivism, Ellis’s defense counsel initially
objected on the grounds that Dr. Branaman did not have a
sufficient factual basis to make a relevant opinion. Following
further questioning of Dr. Branaman by the military judge and
counsel, the defense counsel also objected on the grounds that
“the methodology from which [Dr. Branaman] is basing his opinion
4
The ten risk factors in the Static 99 are: age of offender;
stability of relationships; whether most recent conviction is
for non-sexual violence; any prior conviction for non-sexual
violence; number of sexual offense convictions or charges prior
to most recent offense; has individual been sentenced on more
than four prior occasions; convictions for non-contact sex
offenses; any unrelated victims; any stranger victims; and any
male victims.
6
United States v. Ellis, No. 09-0382/AF
as conducted, does not bear sufficient reliability to be
admissible in this case.” When asked by the military judge why
it was not reliable, defense counsel stated it was because the
test had a seventy to seventy-five percent accuracy rate and
there were cases where the addition of dynamic variables skewed
the accuracy of the Static 99 assessment. Finally, the defense
counsel objected on the grounds that risk of recidivism was not
proper testimony as to rehabilitation potential.
Following this discussion, the military judge stated that
he would allow trial counsel to continue to lay a foundation for
Dr. Branaman’s expert opinion. Trial counsel immediately asked
Dr. Branaman what his opinion was as to Ellis’s risk of
recidivism. Dr. Branaman responded that Ellis fell into the
moderate high category for risk of recidivism, which reflected a
thirty-eight percent chance of recidivism over a fifteen-year
window of time. Dr. Branaman then went on to explain how he
scored each of the ten factors for Ellis, which resulted in a
total score of four points. There was no further ruling on the
admissibility of Dr. Branaman’s expert opinion on Ellis’s risk
of recidivism, nor was there a request for a ruling from either
party.5
5
The defense did object to two further questions for expert
testimony from Dr. Branaman as to Ellis’s treatment amenability
and the potential victim impact on VC. The military judge
sustained both of those objections.
7
United States v. Ellis, No. 09-0382/AF
Discussion
This Court reviews a military judge’s decision to admit or
exclude expert testimony over a defense objection for an abuse
of discretion. United States v. Billings, 61 M.J. 163, 166
(C.A.A.F. 2005). A military judge abuses his discretion when:
(1) the findings of fact upon which he predicates his ruling are
not supported by the evidence of record; (2) if incorrect legal
principles were used; or (3) if his application of the correct
legal principles to the facts is clearly unreasonable. United
States v. Mackie, 66 M.J. 198, 199 (C.A.A.F. 2008). “‘When
judicial action is taken in a discretionary matter, such action
can not be set aside by a reviewing court unless it has a
definite and firm conviction that the court below committed a
clear error of judgment in the conclusion it reached upon
weighing of the relevant factors.’” United States v. Sanchez,
65 M.J. 145, 148 (C.A.A.F. 2007) (quoting United States v.
Houser, 36 M.J. 392, 397 (C.M.A. 1993)).
Ellis argues that Dr. Branaman’s limited record review did
not provide him with a sufficient factual basis for his expert
opinion as to his (Ellis’s) risk of recidivism. Ellis
recognizes that a personal interview is not necessary in order
to give an opinion on the risk of recidivism.6 He argues,
however, that in the absence of a personal interview, Dr.
6
See Stinson, 34 M.J. at 239 (citing Barefoot v. Estelle, 463
U.S. 880, 903-04 (1983); United States v. Hammond, 17 M.J. 218(C.M.A. 1984)).
8
United States v. Ellis, No. 09-0382/AF
Branaman’s preparation was insufficient as he did not review
Ellis’s personnel, mental health, and medical records, nor did
he interview the victims. He also argues that in using the
Static 99 appraisal, Dr. Branaman improperly relied on the
number of charges on the charge sheet and therefore the
probative value of his testimony was marginal and substantially
outweighed by the danger of unfair prejudice under Military Rule
of Evidence (M.R.E.) 403. Since the military judge did not
perform a M.R.E. 403 balancing test on the record, Ellis argues
that his ruling is not entitled to any deference.
The Government responds that Dr. Branaman’s review of the
records and his use of the Static 99 appraisal provided a
sufficient factual basis for his expert opinion. As to Ellis’s
allegation that Dr. Branaman’s appraisal improperly relied on
the number of the charges in the charge sheet, the Government
notes that there was no evidence that the Static 99 appraisal
was unreliable and, in any event, defense counsel effectively
cross-examined Dr. Branaman on the limitations of the Static 99
appraisal when used with discretionary drafting of charges.
Expert Opinion Testimony on Risk of Recidivism
We initially note that Ellis does not challenge whether Dr.
Branaman was qualified as an expert nor does he challenge that
the testimony was within the limits of Dr. Branaman’s expertise.
He bases his challenge on the third Stinson/Banks factor,
arguing that Dr. Branaman did not have a sufficient factual
9
United States v. Ellis, No. 09-0382/AF
basis to provide a relevant expert opinion on his risk of
recidivism.
R.C.M. 1001(b)(5)(A) allows trial counsel to present
opinion evidence as to an accused’s potential for
rehabilitation. R.C.M. 1001(b)(5) goes on to provide, in part:
(B) Foundation for opinion. The witness or deponent
providing opinion evidence regarding the accused’s
rehabilitative potential must possess sufficient
information and knowledge about the accused to offer a
rationally-based opinion that is helpful to the
sentencing authority. Relevant information and
knowledge include, but are not limited to, information
and knowledge about the accused’s character,
performance of duty, moral fiber, determination to be
rehabilitated, and nature and severity of the offense
or offenses.
(C) Bases for opinion. An opinion regarding the
accused’s rehabilitative potential must be based upon
relevant information and knowledge possessed by the
witness or deponent, and must relate to the accused’s
personal circumstances. . . .
Ellis does not assert that the material that Dr. Branaman
did review was neither relevant nor related to Ellis’s personal
circumstances. Rather, he argues that Dr. Branaman should have
reviewed additional materials. The issue here is not whether
Dr. Branaman reviewed every record, it is whether the review he
undertook provided him with “sufficient” information to offer a
rationally based opinion that would be helpful to the sentencing
authority.
There can be no hard and fast rule as to what constitutes
“sufficient information and knowledge about the accused”
necessary for an expert’s opinion as to an accused’s
10
United States v. Ellis, No. 09-0382/AF
rehabilitation potential. In our prior decisions involving the
basis for an expert’s opinion on an accused’s potential for
rehabilitation, we have necessarily analyzed the sufficiency of
the facts and data on a case-by-case basis. United States v.
Gunter, 29 M.J. 140, 141 (C.M.A. 1989) (reviewing data from a
drug rehabilitation file was sufficient basis); Stinson, 34 M.J.
at 235 (reviewing accused confession; observing the guilty plea
inquiry; reviewing the Office of Special Investigation report
and statements by the victim; reviewing the accused’s mental
health records; and interviewing the victim was sufficient
basis); United States v. Scott, 51 M.J. 326, 328 (C.A.A.F. 1999)
(reviewing an accused’s unsworn statement and two mental health
evaluations was sufficient basis); United States v. McElhaney,
54 M.J. 120, 134 (C.A.A.F. 2000) (interviewing the victim and
observations in court were not sufficient basis, also relying on
fact that expert was a child psychiatrist rather than a forensic
psychiatrist).
Dr. Branaman testified that the Static 99 appraisal was
specifically designed to do a risk assessment based upon a
review of records. In addition to the Static 99 assessment, Dr.
Branaman reviewed the charges and specifications, the extensive
stipulation of fact,7 the forensic analysis of the hard drive and
7
The stipulation of fact in this case comprises seventeen pages
thoroughly detailing the underlying circumstances of the
offenses to which Ellis pleaded guilty, including: a
chronological record of how Ellis developed his relationship
11
United States v. Ellis, No. 09-0382/AF
the listing of the images identified there, the chat logs, he
listened to the guilty plea inquiry, and reviewed the
rehabilitation options at Cannon Air Force Base. We conclude
that Dr. Branaman’s review provided a sufficient basis for his
opinion and the military judge did not abuse his discretion in
allowing the testimony.
M.R.E. 403
Ellis goes on to argue that the criteria of the fifth
factor in the Static 99 appraisal, which is based on the number
of the accused’s prior charges and convictions for sexual
offenses, resulted in Dr. Branaman over-relying on the mere
number of charged offenses. As a result of this over-reliance,
Ellis argues that the probative value of the expert testimony
was substantially outweighed by the danger of unfair prejudice
under M.R.E. 403.
Prior to this situation, Ellis had not been charged or
convicted of any sexual offenses. In the instant case Ellis was
charged with eight specifications under Article 134, Uniform
Code of Military Justice (UCMJ), 10 U.S.C. § 934 (2006): two
specifications of indecent acts with VC; one specification of
possession of child pornography; one specification of adultery;
and four specifications of indecent language with VC. In an
with VC; excerpts from the chats between Ellis and VC and Ellis
and “Mandy”; and descriptions of eight video files found on
Ellis’s computers that contained images of children engaged in
sexually explicit conduct.
12
United States v. Ellis, No. 09-0382/AF
additional charge, he was charged under Article 120, UCMJ, 10
U.S.C. § 920 (2006), with carnal knowledge with VC. In a second
additional charge he was charged with two specifications of
attempted indecent language with “Mandy” in violation of Article
134, UCMJ.
Factor five of the Static 99 appraisal takes into account
the sexual offense charges and convictions prior to the “index”
offense. Dr. Branaman explained that the “index” offense is the
last offense committed by the accused, which in this case was
the most recent attempted indecent language charge with “Mandy.”
According to Dr. Branaman, all of the other charged offenses
became “prior” charges for Static 99 purposes. Under the Static
99 criteria for factor five, these “prior” charges resulted in
Ellis receiving three points. At the time of the appraisal,
Ellis had not been convicted of any sexual offenses and he was
therefore given zero points for prior convictions. This
resulted in a score of three points for factor five. Ellis also
received an additional point in factor eight because the victims
were not related to him. Ellis’s total score under the Static
99 was four points, which placed him in the moderate high risk
category.
Ellis argues that if he had been charged with one indecent
language offense on divers occasions with VC, the number of
prior charges against him would have been reduced by three,
which would have reduced his score for factor five, resulting in
13
United States v. Ellis, No. 09-0382/AF
a lower total score and a lower risk category. Because of the
methodology of factor five, Ellis argues that since “the
principal basis for Dr. Branaman’s expert opinion relied so
heavily on the mere form and number of charged offenses . . .
his testimony did not aid the military judge in determining
appellant’s rehabilitative potential.” As a result, Ellis
argues that the probative value of Dr. Branaman’s expert
testimony based on the appraisal is substantially outweighed by
the danger of unfair prejudice, citing M.R.E. 403.
We initially note that the four indecent language
specifications involving VC all involved different conduct and
occurrences. While Ellis did object at trial as to the
reliability of Dr. Branaman’s methodology, that objection was
limited to the seventy to seventy-five percent accuracy rate of
the assessment and the assertion that the addition of dynamic
variables could skew the accuracy of the assessment. Ellis did
not object to the methodology of factor five, nor did he request
a Daubert/Houser inquiry as to the validity of the Static 99
instrument.8 At oral argument before this court Appellant’s
counsel clarified that Ellis was not challenging the validity of
the Static 99 appraisal. Without such a challenge it is
8
See Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579(1993); United States v. Houser, 36 M.J. 392 (C.M.A. 1993),
cert. denied, 510 U.S. 864 (1993). Although Static 99 is widely
used, the issue as to whether it would meet the Daubert standard
is the subject of ongoing judicial debate. See Judge Posner’s
critical discussion in McIlrath, 512 F.3d at 425.
14
United States v. Ellis, No. 09-0382/AF
difficult for Ellis to argue that an expert opinion based on the
Static 99 appraisal should not have been admitted under M.R.E.
403 because the methodology under factor five may have been
flawed.
Once the Static 99 appraisal was accepted by the military
judge, the challenges that Ellis had as to methodology went to
weight rather than admissibility. In Barefoot, 463 U.S. at 901,
the Supreme Court discussed the usefulness of psychiatric
predictions of future dangerousness given the number of studies
that indicated those predictions were often inaccurate. In
allowing the testimony the Court stated that it was unconvinced
“that the adversary process cannot be trusted to sort out the
reliable from the unreliable evidence, particularly when the
convicted felon has the opportunity to present his own side of
the case.” Id.Both the defense counsel and the military judge thoroughly
questioned Dr. Branaman concerning the methodology of the Static
99 appraisal. Ellis’s defense counsel specifically cross-
examined Dr. Branaman about the effect of the Government’s
charging decisions in relation to factor five of the Static 99
appraisal. As a result of this cross-examination, Dr. Branaman
acknowledged that had the indecent language charges involving VC
been consolidated, Ellis may have been placed in a lower risk
category.
15
United States v. Ellis, No. 09-0382/AF
The military judge was therefore aware of the issue of
potential Government influence as a result of their charging
decisions on factor five of the Static 99 appraisal, and, as the
trier of fact in this case, is presumed to have given it
appropriate weight. United States v. Robbins, 52 M.J. 455, 457
(C.A.A.F. 2000) (A military judge is presumed to know the law
and apply it correctly, is presumed capable of filtering out
inadmissible evidence, and is presumed not to have relied on
such evidence on the question of guilty or innocence.).
DECISION
Under the circumstances of this case, the military judge
did not abuse his discretion in allowing Dr. Branaman’s expert
testimony as to Ellis’s risk of recidivism. The decision of the
United States Air Force Court of Criminal Appeals is affirmed.
16
United States v. Ellis, No. 09-0382/AF
BAKER, Judge (concurring in the result):
I write separately to emphasize that the result in this
case is limited to the facts of this case. Among other things,
this was a military judge alone sentencing proceeding and
Appellant did not object to the admission of the Static 99
information on Daubert1 grounds.
The majority opinion identifies these points. What it does
not say is that the Static 99 system has not been fully vetted
under Daubert in the context of the military justice system. In
the civilian context, at least one circuit court of appeals has
expressed skepticism regarding the system. In that case, Judge
Posner stated:
[Static 99] may be more accurate than clinical
assessments . . . but that may not be saying much.
Estimates of recidivism are bound to be too low when
one is dealing with underreported crimes such as sex
offenses. Static 99 treats as a recidivist only
someone who is convicted of a further sex offense, but
the recidivism concern is with someone who commits a
further offense, whether or not he is caught -- yet if
he is not caught, his subsequent crime does not affect
the data on which the Static 99 calibrations are
based.
United States v. McIlrath, 512 F.3d 421, 425 (7th Cir. 2008).
My concern is with how the Static 99 system should be used in
military sentencing, if at all.
1
Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579(1993).
United States v. Ellis, No. 09-0382/AF
Military sentencing is predicated on the individualized
consideration of the accused. This is well established in our
case law. See United States v. Mamaluy, 10 C.M.A. 102, 106, 27
C.M.R. 176, 180 (1959) (“accused persons are not robots to be
sentenced by fixed formulae but rather, they are offenders who
should be given individualized consideration on punishment”);
see also United States v. Snelling, 14 M.J. 267, 268 (C.M.A.
1982). A formulaic methodology used for sentencing such as the
Static 99 used here would seem to convert individualized
consideration into a numeric calculation based on static
factors, including matters that in the military justice system
are inherently discretionary, like whether the prosecutor
charges conduct “on divers occasions” or through multiple
counts.
Nonetheless, in this case, I am confident Appellant
received the individualized sentencing consideration the Uniform
Code of Military Justice requires. First, this was a military
judge alone trial. A military judge in particular is suited to
understand the difference between a statement that thirty-eight
percent of persons with the same Static 99 score re-offended and
a statement that there was a thirty-eight percent chance that
this Appellant would re-offend. Second, Appellant had ample
opportunity to cross-examine the expert and place his concerns
and doubts before the military judge. Third, the seventeen-page
2
United States v. Ellis, No. 09-0382/AF
stipulation of fact put the static factors on which the system
relies into the individualized context of this case.
3