United States v. Mullins

Judge BAKER delivered the opinion of the Court. A general court-martial composed of members convicted Appellant, contrary to his pleas, of rape of a child, forced sodomy of a child, two specifications of indecent acts and two specifications of possession of child pornography, in violation of Articles 120, 125, and 134, Uniform Code of Military Justice (UCMJ), 10 U.S.C. §§ 920, 925, and 934 (2000), respectively. The adjudged and approved sentence included a dishonorable discharge, confinement for ten years, forfeiture of all pay and allowances, and reduction to the grade of E-l. The United States Navy-Marine Corps Court of Criminal Appeals (CCA) dismissed one of the specifications of child pornography, but affirmed the remaining findings of guilt and the sentence. United States v. Mullins, No. NMCCA 200200988, 2006 CCA LEXIS 327, at *46, *115 2006 WL 4573011, at *16 (N.M.Ct.Crim.App. Dec. 7, 2006) (unpublished). In 2008, this Court granted review of two issues, including one regarding expert testimony on the frequency of false positives in cases of child molestation. The Court set aside the decision of the CCA and remanded for a new Article 66(c), UCMJ, 10 U.S.C. § 866(c) (2006), review. United States v. Mullins, 66 M.J. 468 (C.A.A.F.2008). In the second CCA opinion, the court held that while there was error in allowing the expert to testify about the children’s veracity, for the purposes of plain error review it was neither obvious nor prejudicial in light of the military judge’s instructions. United States v. Mullins, No. NMCCA 200200988, 2009 CCA LEXIS 171, at *15, 2009 WL 1393229, at *6 (N.M.Ct.Crim.App. May 14, 2009) (unpublished). The CCA adopted the other conclusions from the first CCA opinion. Id. at *22-*23, 2009 WL 1393229, at *8. We granted review of the following issues: I. WHETHER THE LOWER COURT ERRED IN HOLDING THAT THERE WAS NOTHING IMPERMISSIBLE IN THE MILITARY JUDGE ALLOWING THE GOVERNMENT TO INTRODUCE LIE DETECTOR TESTIMONY IN VIOLATION OF MILITARY RULE OF EVIDENCE 702. II. WHETHER THE LOWER COURT DENIED APPELLANT DUE PROCESS WHEN IT DENIED HIM RELIEF DUE TO EXCESSIVE POST-TRIAL PROCESSING DELAY AND DENIED HIS SUPPLEMENTAL ASSIGNMENTS OF ERROR. We hold that it was error to admit expert testimony from which members could infer there was a 1 in 200 chance that the allegations were false. However, we conclude that the error did not materially prejudice Appellant’s substantial rights in light of the military judge’s corrective instructions and the time at which they occurred. Additionally, Appellant’s due process rights were not violated because the post-trial delay in this case was not prejudicial. I. EXPERT TESTIMONY A Background Appellant had two daughters with Tiffany Miller, DM and SM, who were nine years old and seven years old respectively, at the time of the offenses. On June 18, 2000, DM told her mother that Appellant had done “rude things” to her. A few days later, both girls were interviewed by a forensic specialist and a few weeks later they were examined by a sexual assault nurse examiner. DM and SM testified that between June 1999 and January 2000, Appellant forced them to perform indecent acts on him, including oral sex and masturbation. SM testified that she had been raped. The girls also stated that Appellant had child pornography on his computer and forced them to watch those materials. During the trial, Cynthia Conrad, a forensic child interviewer for the Kitsap County prosecutor’s office, testified about the types of interviews she performs. She stated that a normal seven- to nine-year-old child might understand sexual intercourse but would not understand oral or anal sex, male masturbation, or ejaculation. She also testified that the characteristics she saw in the victims’ interviews were “consistent ... with a child who had been sexually abused or ... a child who may have been sexually abused.” In response to her testimony, the military judge gave a sua sponte instruction, stating: [N]o witness is a human lie detector. That is no one — no one who testifies in this courtroom can know if someone else is telling the truth or lying. You are advised that only you, the members of this court, can determine the credibility of the witnesses and what the ultimate facts of this case are. No witness, including an expert witness, can testify that someone else’s account of what happened is true or credible, that a person believes the alleged victim or that, in fact, a sexual encounter actually occurred. On redirect, Ms. Conrad testified about the frequency of children lying about sexual abuse, saying that it was less than “1 out of 100 or 1 out of 200.” The military judge then asked Ms. Conrad: *116 [D]o you have any forensic, that is, scientifically accurate way of proving whether the child is telling the truth or not? In other words ... the only way that you typically could know that is if the child later comes forth and says ‘Yes, I made it up,’ or ... unless that [defendant] ultimately confesses, you would ultimately never know who was telling the truth and who wasn’t, is that correct? Ms. Conrad responded affirmatively: “That’s correct.” There was no objection at trial and defense counsel cited this last bit of testimony during his closing argument. B. Analysis Appellant argues that the military judge erred by admitting testimony from Ms. Conrad about the frequency with which children make false claims of sexual abuse. Appellant argues that the members might infer from the expert’s testimony about children generally an equivalent situation in Appellant’s case, i.e., that there was a 1 in 200 chance that Appellant was innocent. As a result, Appellant contends that the military judge should have provided an immediate corrective instruction to the members and struck the testimony from evidence, so that he would not be materially prejudiced. “Where an appellant has not preserved an objection to evidence by making a timely objection, that error will be forfeited in the absence of plain error.” United States v. Brooks, 64 M.J. 325, 328 (C.A.A.F.2007) (citing Military Rule of Evidence (M.R.E.) 103(d)). In this case, defense counsel did not object to Ms. Conrad’s testimony during the trial. The plain error standard is met when “(1) there is error, (2) the error is plain or obvious, and (3) the error results in material prejudice to a substantial right of the accused.” United States v. Fletcher, 62 M.J. 175, 179 (C.A.A.F.2005) (citing United States v. Rodriguez, 60 M.J. 87, 88-89 (C.A.A.F.2004)); United States v. Robbins, 52 M.J. 455, 457 (C.A.A.F.2000). “Our standard of review for determining whether there is plain error is de novo.” Brooks, 64 M.J. at 328 (citing United States v. Gudmundson, 57 M.J. 493, 495 (C.A.A.F.2002)). 1. The error in this case In a trial involving the sexual assault of a child, “‘[a]n expert may testify as to what symptoms are found among children who have suffered sexual abuse and whether the child-witness has exhibited these symptoms.’” United States v. Birdsall, 47 M.J. 404, 409 (C.A.A.F.1998) (quoting United States v. Harrison, 31 M.J. 330, 332 (C.M.A.1990)). “However, an expert may not testify regarding the credibility or believability of a victim, or ‘opine as to the guilt or innocence of an accused.’ ” United States v. Cacy, 43 M.J. 214, 217 (C.A.A.F.1995) (quoting United States v. Suarez, 35 M.J. 374, 376 (C.M.A.1992)); see also Brooks, 64 M.J. at 328 & nn. 2-3. This ease is similar to Brooks. There an expert witness testified that the frequency of false sexual abuse allegations was approximately five percent. Brooks, 64 M.J. at 327. This Court concluded that such testimony was “the functional equivalent of saying that the victim in a given case is truthful and should be believed” and held that the military judge erred by admitting it. Id. at 329. The testimony in this ease also involves a statistical statement of how often false accusations of sexual abuse occur, raising the risk that the members would infer an equivalent likelihood in Appellant’s case. Such an inference derived from expert testimony would invade the province of the court members to determine the credibility of witnesses. Our conclusion is that it was error to admit the statistical testimony in Appellant’s case. An expert inference that there is a 1 in 200 chance the victim is lying undermines the duty of the panel members to determine guilt beyond a reasonable doubt. 2. The error was plain and obvious There are several reasons supporting our determination that the error was plain and obvious in this case. First, on direct review, we apply the clear law at the time of appeal, not the time of trial. United States v. Harcrow, 66 M.J. 154, 159 (C.A.A.F.2008) (citing Johnson v. United States, 520 U.S. 461, 468, 117 S.Ct. 1544, 137 L.Ed.2d *117 718 (1997)). This case was at the CCA when Brooks, a case holding that expert testimony about the statistical frequency of children lying about sexual abuse is inadmissible, was decided. 64 M.J. at 328-30. In Brooks, we concluded there was plain error. Therefore, it follows that an error that was plain and obvious in Brooks would be plain and obvious in a subsequent case when there were no intervening changes in the law. Second, related ease law at the time of trial also supports the conclusion that the error in this case was plain and obvious. In United States v. Banks, for example, this Court cautioned against expert statistical testimony that placed an accused within a definitive profile suggesting guilt. 36 M.J. 150, 161-63 (C.M.A.1992). Although after Appellant’s trial, United States v. Traum echoed this concern. We reversed, finding that the expert’s “statement placed a statistical probability on the likelihood that Appellant committed the offense.” 60 M.J. 226, 235-36 (C.A.A.F.2004). Third, the error in this case was apparent to the military judge. This is evident in the military judge’s sound decision to immediately issue a corrective instruction on the role of members when the expert initially stated that the children’s statements were consistent with those of children who had been abused. He reiterated this instruction, in generic form, before the members recessed for deliberations. He also asked a clarifying question directly after the problematic testimony. Thus, while the military judge’s action in addressing the testimony was commendable, it also supports the conclusion that the erroneous nature of the testimony was obvious to him at the time. The question is whether these remedial steps were sufficient to cure any potential prejudice arising from the statistical statement. 3. Prejudice The last step in plain error analysis is to test whether an error materially prejudiced Appellant. Prejudice results when there is “undue influence on a jury’s role in determining the ultimate facts in the case.” Birdsall, 47 M.J. at 411. We look at the erroneous testimony in context to determine if the witness’s opinions amount to prejudicial error. United States v. Eggen, 51 M.J. 159, 161 (C.A.A.F.1999). Context includes such factors as the immediate instruction, the standard instruction, the military judge’s question, and the strength of the government’s ease — to determine whether there was prejudice. “Absent evidence to the contrary, court members are presumed to comply with the Military Judge’s instructions.” United States v. Thompkins, 58 M.J. 43, 47 (C.A.A.F.2003). Here, the military judge gave an instruction at the end of Ms. Conrad’s direct examination, as well as before deliberations. The timing of these instructions distinguishes this case from Brooks, where the military judge only instructed the panel before the members deliberated, a fact noted and relied upon by this Court. 64 M.J. at 330. Here, the military judge gave an instruction on credibility, ensuring that the panel members would know their role and not accept the percentage testimony as a proxy for credibility. We also find it hard to fault the military judge for not repeating the same instruction shortly after he gave it the first time. 1 If the members complied with the instructions then Ms. Conrad’s testimony should not have inappropriately bolstered the victims’ credibility. The military judge also asked Ms. Conrad a clarifying question, the answer to which indicated that she did not have “a scientifically accurate way of proving whether [a] child is telling the truth or not,” thus minimizing the impact of her testimony. Because of the military judge’s questions, the CCA found that the testimony was based on the expert’s personal experience, instead of scientific studies. As a result, the testimony did not carry the same weight with the panel members as the testimony offered in Brooks. Thus, while Appellant is correct that a judi *118 cial question is not the same as a corrective instruction, we are hard-pressed not to conclude that, given the timing of the first instruction as well as the question and subsequent answer, the taint from the statistical evidence was cured. Appellant argues that the testimony was prejudicial because it supplemented and buttressed a weak case. As in Brooks, the Government had “no other direct witnesses, no confession, and no physical evidence to corroborate the victim’s sometimes inconsistent testimony.” Brooks, 64 M.J. at 330. However, here there was corroborating evidence upon which the court members could rely. Both victims testified and were fully cross-examined. Non-relative witnesses testified about the fear the girls had of their father. The victims’ testimony was supported by the presence of child pornography, illicit instant message chat sessions found on Appellant’s computer, and the properly admitted testimony of the expert witness. In short, the members had other reasons to believe the victims. Therefore, in this case, we conclude there was sufficient other evidence and the members were properly instructed such that we are convinced that they were able to come to a decision in the case without relying on any credibility determinations offered by Ms. Conrad. Appellant has failed to demonstrate prejudice. II. POST-TRIAL DELAY AND DUE PROCESS A. Background Appellant’s trial was completed on April 6, 2001, but the convening authority’s action did not occur until April 4, 2002, over 360 days later. Then 448 days elapsed between the date the record was docketed with the CCA and the date of the first appointed appellate defense counsel’s initial contact with Appellant. Appellant had, in succession, four separate appointed appellate attorneys. He filed various writs and motions pro se, including complaints about delay in the appellate process. Appellant was released from confinement on March 9, 2007, and was immediately placed on appellate leave status. According to the appellate record, he then applied for unemployment insurance from the state of California on May 4, 2007. A few days later he received notice from the California unemployment office that it could not give him benefits because he was still on appellate leave status and had not received a DD-214. B. Analysis “We review de novo claims that an appellant has been denied the due process right to a speedy posNtrial review and appeal.” United States v. Moreno, 63 M.J. 129, 135 (C.A.A.F.2006); United States v. Rodriguez, 60 M.J. 239, 246 (C.A.A.F.2004). When considering appellate delay, a court must balance four factors: “(1) the length of the delay; (2) the reasons for the delay; (3) the appellant’s assertion of the right to timely review and appeal; and (4) prejudice.” Moreno, 63 M.J. at 135 (citing Barker v. Wingo, 407 U.S. 514, 530, 92 S.Ct. 2182, 33 L.Ed.2d 101 (1972)). Where an appellant meets his burden in demonstrating unreasonable appellate delay, the burden shifts to the government to show that the due process violation was harmless beyond a reasonable doubt. United States v. Ashby, 68 M.J. 108, 125 (C.A.A.F.2009). Even assuming a due process violation occurred in this case, 2 we hold that it was harmless beyond a reasonable doubt because the record contains no evidence of prejudice warranting relief. In examining the Government’s burden to show harmlessness beyond a reasonable doubt, we review Appellant’s three arguments regarding prejudice: the delay kept him from receiving unemployment benefits because he lacked a DD-214, it increased his anxiety because he had to register as a sex offender, and “[a] more timely appeal ... would have enabled him to initiate legal proceedings to obtain visitation and legal custody of his now-grown children.” Since *119 Appellant has not prevailed on the expert testimony issue, he cannot claim that the delay hurt his ability to retry the case or would have enabled him to seek custody of his children in a more timely fashion. Moreno, 63 M.J. at 140. The question of unemployment benefits is a closer call. Appellant asserts that if his appeal had been adjudicated with less delay he would no longer be on appellate leave and would have received unemployment benefits. Appellant argues this is analogous to “recognized interference with post-military employment opportunities as a form of prejudice that warrants relief for unreasonable post-trial delay.” United States v. Jones, 61 M.J. 80, 84 (C.A.A.F.2005) (citing United States v. Sutton, 15 M.J. 235 (C.M.A.1983)). The appellate question is not whether such a scenario could amount to prejudice; it could. The question is whether the record reflects that such prejudice existed in this case. We conclude that the record does not demonstrate that it was Appellant’s leave status that kept him from receiving the benefits and that he would not have been denied on some other grounds. 3 See Bush, 68 M.J. at 103 n. 8. In United States v. Schweitzer, 68 M.J. 133, 138-39 (C.A.A.F.2009), and Ashby, 68 M.J. at 125, this Court denied the appellants’ assertions that post-trial delay was prejudicial because it caused difficulty in finding adequate employment. In Ashby, as in Jones, the Court was provided with affidavits from would-be employers supporting the appellants’ claims. 68 M.J. at 125 n. 11 (Ashby); 61 M.J. at 81 (Jones). The record in this case does not contain an equivalent affidavit, nor does it contain any other authoritative evidence that a person in Appellant’s circumstances would have been eligible for unemployment benefits and received them once his appeal was final. Having carefully examined the entire record and finding no convincing evidence of prejudice, we conclude that, under the totality of the circumstances, the post-trial delay was harmless beyond a reasonable doubt. CONCLUSION The decision of the United States Navy-Marine Corps Court of Criminal Appeals is affirmed. Footnotes [1] The record of trial has only seven pages of testimony between the military judge's first instruction and the expert’s statistical testimony (during which time there was only a six-minute recess). [2] " '[No] single factor is required for finding a due process violation and the absence of a given factor will not prevent such a finding.’ ” United States v. Bush, 68 M.J. 96, 103 n. 8 (C.A.A.F.2009) (quoting Moreno, 63 M.J. at 136). [3] Appellant filed two motions with this Court on April 16, 2010, that were denied. One was to take judicial notice of a sixteen-page printout from the California Employment Development Department addressing employment benefits and misconduct generally. The other motion was to attach a California Unemployment Insurance Program fact sheet and the Appellant's most recent Social Security statement. Neither directly addressed Appellant's circumstances.