United States v. Eslinger

UNITED STATES, Appellee/Cross-Appellant v. Alan D. ESLINGER, Sergeant First Class U.S. Army, Appellant/Cross-Appellee Nos. 10-0537 and 11-5002 Crim. App. No. 20070335 United States Court of Appeals for the Armed Forces Argued January 24, 2011 Decided July 8, 2011 BAKER, J., delivered the opinion of the Court, in which STUCKY and RYAN, JJ., joined. ERDMANN, J., filed a separate dissenting opinion, in which EFFRON, C.J., joined. Counsel For Appellant/Cross-Appellee: Captain Matthew T. Grady (argued); Colonel Mark Tellitocci, Lieutenant Colonel Jonathan F. Potter, Major Laura R. Kesler, and Captain Jess B. Roberts (on brief); Captain Jennifer A. Parker. For Appellee/Cross-Appellant: Captain Joshua W. Johnson (argued); Colonel Michael E. Mulligan, Major Christopher B. Burgess, and Major Adam S. Kazin (on brief). Military Judge: Michael J. Hargis THIS OPINION IS SUBJECT TO REVISION BEFORE FINAL PUBLICATION. United States v. Eslinger, Nos. 10-0537/AR and 11-5002/AR Judge BAKER delivered the opinion of the Court. A general court-martial composed of officer and enlisted members convicted Appellant/Cross-Appellee (Appellant), contrary to his pleas, of three specifications of possession of child pornography, in violation of Article 134, Uniform Code of Military Justice (UCMJ), 10 U.S.C. § 934 (2006). The adjudged and approved sentence included confinement for three years, reduction to pay grade E-1, forfeiture of all pay and allowances, and a bad-conduct discharge. On review, the United States Army Court of Criminal Appeals affirmed.1 We granted review of the following issue: WHETHER THE MILITARY JUDGE COMMITTED PLAIN ERROR BY PERMITTING THE GOVERNMENT TO OFFER EVIDENCE IN THE FORM OF OPINION TESTIMONY FROM SENIOR OFFICER AND NCO WITNESSES WITH NO PERSONAL KNOWLEDGE OF APPELLANT’S DUTY PERFORMANCE TO OPINE THAT HE SHOULD BE SEPARATED FROM THE ARMY AND SPECIAL FORCES. In addition, the Government certified the following issue: WHETHER THE RESTRICTIONS UNDER R.C.M. 1001(b)(5) APPLY TO REBUTTAL EVIDENCE SUBMITTED UNDER R.C.M. 1001(d) AND WHETHER THE U.S. ARMY COURT OF CRIMINAL APPEALS ERRED IN FINDING THAT THE MILITARY JUDGE COMMIT[T]ED PLAIN AND OBVIOUS ERROR WHEN HE PERMITTED INTRODUCTION OF GOVERNMENT REBUTTAL TESTIMONY TO DEFENSE “RETENTION EVIDENCE” WHEN THERE WAS NO DEFENSE OBJECTION. 1 United States v. Eslinger, No. ARMY 20070335, 2010 CCA LEXIS 64, at *48 (A. Ct. Crim. App. May 21, 2010). 2 United States v. Eslinger, Nos. 10-0537/AR and 11-5002/AR In our view the legal questions at the root of these issues were addressed in United States v. Griggs, 61 M.J. 402 (C.A.A.F. 2005). Thus the question presented here is how the Griggs analysis applies to these facts. For the reasons stated below we conclude that if any errors with regard to a lack of foundation for lay opinions were clear and obvious, they did not substantially influence the adjudged sentence. BACKGROUND Appellant was a Special Forces medic who had served in the Army for eighteen years. In March 2006, Appellant had been living with his girlfriend Loren R. Masden for two years. She and Appellant exchanged various computer passwords as a sign of trust. On March 4, 2006, while Appellant was at “Pinon Canyon training site in Trinidad,” Masden logged onto Appellant’s laptop and discovered images of child pornography on the computer. She immediately called her sister, who came to Masden and Appellant’s home and also saw the images. Masden testified that her sister saw a digital fingerprint indicating that the pictures were downloaded on February 14, 2006, a time at which Appellant had been in North Carolina for training. Masden was upset and went to stay at her sister’s, returning a few days later to pack her things. On March 8, 2006, she reported the images to law enforcement. 3 United States v. Eslinger, Nos. 10-0537/AR and 11-5002/AR Appellant testified that at the time the images were downloaded, he was in training and occasionally did not have possession of his laptop or remote storage drives, and that some of the images may have been downloaded unintentionally while intending to download adult pornography and other materials from a file-sharing website. There were over 1,700 pornographic images found on Appellant’s computer that depicted children. These images included various forms of child pornography including anal, oral, and vaginal penetration of children under the age of two. The members found Appellant guilty of the charged offenses. At sentencing Appellant introduced testimony from three mitigation witnesses, each of whom basically testified that in his opinion, Appellant should be retained in the armed forces. Master Sergeant (MSG) Willie D. Gibbons, a member of 3d Battalion, 10th Special Forces Group, stated among other things, “I’ve already packed his bags . . . I would take him on my team in a minute,” and “Just like an alcoholic . . . I think, you know, something needs to be done . . . . Past that, I think he needs to stay in the service.” Captain Timothy J. Coffman, the battalion physician assistant, stated, “He is my best medic” and “I think we should rehabilitate him . . . . I mean, he’s a great soldier. He has a great service record as far as military activities.” Sergeant First Class Shawn Dishman, a member of 4 United States v. Eslinger, Nos. 10-0537/AR and 11-5002/AR Appellant’s company, whose testimony was admitted through a stipulation of expected testimony stated: I definitely think that there is a place for [Appellant] in the Army and within the 10th Special Forces. I truly believe that Special Forces is the only place for SFC Eslinger. I would be proud to serve with him in the future despite this conviction. . . . [I] would welcome him to my team any day. In rebuttal to Appellant’s mitigation evidence, the Government introduced testimony from five witnesses. Major (MAJ) Isaac J. Peltier, the acting battalion commander of 3rd Battalion, 10th Special Forces Group, Appellant’s battalion, stated: It is my opinion that, clearly [Appellant] should not deploy to combat with this organization. . . . And for that matter, he should not return to this -- the 3rd Battalion. And I’ll go a step further in my opinion, based on his pattern of misconduct, he shouldn’t even be in the Army. Sergeant Major (SGM) Jason M. Krider, Appellant’s battalion command sergeant major (CSM), testified, “There is no place in our ranks for Sergeant Eslinger.” MSG Timothy D. Stensgaard, one of the two team sergeants of the tactical support detachment in the 10th Special Forces Group testified, “As a leader in the United States Army, I don’t feel that based on his prior incidences and this conviction how he [sic] could remain in the U.S. Army and effectively serve.” Colonel (COL) Kenneth E. 5 United States v. Eslinger, Nos. 10-0537/AR and 11-5002/AR Tovo, group commander of the 10th Special Forces Group, testified: Sergeant Eslinger’s got a good reputation as a soldier, particularly a combat soldier, in the Group. However . . . . [Y]ou just listed four fairly significant instances of ill-discipline, and frankly, that’s more chances than we allow a guy. . . . I just find that . . . his ill-discipline is incompatible with continued service, certainly within the 10th Group. CSM Charles M. Sekelsky, the group command sergeant major, testified, “I think he’s embarrassed the regiment and the United States Army for his actions.”2 In addition, on cross-examination defense counsel elicited agreement from the sergeant major that Appellant was an exceptional medic and an exceptional team member when deployed in a combat zone. Defense counsel specifically asked CSM Sekelsky, “If you could put him in a can and take him to Iraq and only open him up in Iraq, you’d prefer to do it that way, wouldn’t you?” To which he responded, “Yes.” With two exceptions, defense counsel did not object to the testimony of these witnesses. Following SGM Krider’s testimony the military judge asked, “Any issues with the sergeant major’s testimony?” Defense counsel, citing Rule for Courts-Martial (R.C.M.) 1001(b)(5), responded with a request that the military judge instruct the members to disregard the testimony because the offenses of which Appellant was found guilty formed the 2 CSM Sekelsky had also been Appellant’s previous battalion command sergeant major. 6 United States v. Eslinger, Nos. 10-0537/AR and 11-5002/AR principal basis for SGM Krider’s opinion. Following an Article 39(a), UCMJ, 10 U.S.C. § 839(a) (2006), session to discuss the matter, the military judge instructed the members to disregard SGM Krider’s testimony. In addition, prior to CSM Sekelsky’s testimony the military judge asked trial counsel for an offer of proof as to what the witness would testify. Defense counsel objected to CSM Sekelsky’s expected testimony as cumulative since the members had just heard from the Group commander, COL Tovo. In counsel’s view, the colonel had already testified as to the consensus of the command. The military judge overruled this specific defense objection noting that CSM Sekelsky “appears to have some closer connection with the accused.” Immediately after ruling on this objection to the witness’s testimony, the military judge asked defense counsel specifically, “Any other objections, defense?” Defense counsel responded, “No, sir.” At the close of the sentencing case, although the defense requested a number of specific instructions they did not request a specific instruction regarding the Government’s rebuttal evidence. With respect to the sentencing evidence, the military judge provided the standard instruction to “consider all matters in extenuation and mitigation as well as those in aggravation.” He also instructed the members to consider, among other things, 7 United States v. Eslinger, Nos. 10-0537/AR and 11-5002/AR evidence of Appellant’s good military character, his combat record, and his record in the service for bravery. Before the Court of Criminal Appeals Appellant challenged the admission of the Government’s rebuttal evidence on the ground that the witnesses did not have adequate foundations to provide their opinions as to whether Appellant should be retained in the armed forces. Eslinger, 2010 CCA LEXIS 64, at *2. The lower court, sitting en banc and without distinguishing between the various rebuttal witnesses, found “clear and obvious error in the admission of evidence which both lacked foundation and raised command influence concerns, without proper limiting instruction[s].” Id. at *42. However, after a detailed review for prejudice, that court concluded that any errors were harmless. Id. at *46-*47. Before this Court Appellant renews his argument that the military judge committed plain error in allowing the testimony of the rebuttal witnesses. First, Appellant argues the witnesses lacked an adequate foundation to form and offer an opinion on retention. Second, he argues that the restrictions on evidence of rehabilitative potential contained in R.C.M. 1001(b)(5) apply to Government rebuttal as well as to the Government’s sentencing case-in-chief. Finally, Appellant challenges the lower court’s conclusion that any errors were harmless. For its part, the Government challenges the lower 8 United States v. Eslinger, Nos. 10-0537/AR and 11-5002/AR court’s conclusion that the limitations in R.C.M. 1001(b)(5) apply to otherwise properly admitted rebuttal evidence. DISCUSSION A military judge’s decisions to admit or exclude evidence are reviewed for an abuse of discretion. United States v. Ediger, 68 M.J. 243, 248 (C.A.A.F. 2010). Failure to object to the admission of evidence at trial forfeits appellate review of the issue absent plain error. United States v. Kasper, 58 M.J. 314, 318 (C.A.A.F. 2003) (citation omitted); United States v. Halford, 50 M.J. 402, 404 (C.A.A.F. 1999); United States v. Raya, 45 M.J. 251, 253 (C.A.A.F. 1996). R.C.M. 1001(b)(5) and Rebuttal Evidence We begin our analysis with the threshold question as to whether R.C.M. 1001(b)(5) bars the Government from introducing testimony of the type and quality in this case in rebuttal to defense retention evidence. If so, there is no need to consider whether there was a proper foundation for doing so. Evidence that goes toward the accused’s rehabilitative potential is permissible at sentencing. “The trial counsel may present, by testimony or oral deposition in accordance with R.C.M. 702(g)(1), evidence in the form of opinions concerning the accused’s previous performance as a servicemember and potential for rehabilitation.” R.C.M. 1001(b)(5)(A). However, “A witness may not offer an opinion regarding the 9 United States v. Eslinger, Nos. 10-0537/AR and 11-5002/AR appropriateness of a punitive discharge or whether the accused should be returned to the accused’s unit.” R.C.M. 1001(b)(5)(D). In United States v. Ohrt, 28 M.J. 301 (C.M.A. 1989), the Court concluded that this restriction applies to both government and defense sentencing evidence. “[A] witness -- be he for the prosecution or the defense -- should not be allowed to express an opinion whether an accused should be punitively discharged.” Id. at 304-05. “[A]ppropriateness of punishment” is an issue to be decided by the members and “cannot be usurped by a witness.” Id. at 305. However, in Griggs, we held that “R.C.M. 1001(b)(5)(D) does not apply to defense mitigation evidence, and specifically does not preclude evidence that a witness would willingly serve with the accused again.” 61 M.J. at 409. This conclusion was based in part on the fact that “so-called ‘retention evidence’ is classic matter in mitigation, which is expressly permitted to be presented by the defense.” Id. However, we reached this conclusion with three important cautionary caveats. First, “there can be a thin line between an opinion that an accused should be returned to duty and the expression of an opinion regarding the appropriateness of a punitive discharge.” Id. Second, concerns raised with respect to this distinction “can be addressed with a tailored instruction focusing on the distinction between a punitive discharge, which is for the 10 United States v. Eslinger, Nos. 10-0537/AR and 11-5002/AR members to decide, and the willingness of a servicemember to serve with an accused again.” Id. Third, and most importantly for the purpose of this case, we directly responded to the Government’s argument in Griggs that if the defense were allowed to admit such evidence the Government would be without recourse. We stated: Consistent with the historical concerns regarding command influence, the Government is free to rebut such assertions. As stated in [United States v.] Aurich,[3] “if an accused ‘opens the door’ by bringing witnesses before the court who testify that they want him or her back in the unit, the Government is permitted to prove that that is not a consensus view of the command.” Id. at 410 (footnote added). We continue to adhere to this view. As in other contexts, where a party opens the door, principles of fairness warrant the opportunity for the opposing party to respond, provided the response is fair and is predicated on a proper testimonial foundation. See United States v. Blau, 5 C.M.A. 232, 244, 17 C.M.R. 232, 244 (1954) (otherwise “an accused would occupy the unique position of being able to ‘parade a series of partisan witnesses before the court’ . . . without the slightest apprehension of contradiction or refutation”). In this case, the defense counsel opened the door to rebuttal through testimony from its witnesses indicating that they would gladly serve with Appellant again. Therefore the 3 31 M.J. 95, 96-97 (C.M.A. 1990). 11 United States v. Eslinger, Nos. 10-0537/AR and 11-5002/AR Government was free to rebut with proper evidence that this was not the consensus of the command. Three concerns warrant emphasis. First, when the government’s evidence in rebuttal to defense retention evidence is testimony of the accused’s commander, it may well “raise the specter of command influence.” Griggs, 61 M.J. at 408 (citation omitted). However, we hasten to note that evidence that the defense witnesses’ views are “not a consensus view of the command” simply means that retaining the accused is not the view of every member of the command. See id. at 410 (citation and quotation marks omitted). It does not necessarily mean that the government may parade the commanding officer and the rest of the accused’s chain of command to have them give a command view on retention. That would depend on just how wide the defense opened the door. Second, to be clear, a commander may testify, but it is essential for the military judge to be on guard for the possibility, intended or not, that a commander’s testimony could convey undue command influence to the members. While not an absolute requirement, a tailored instruction from the military judge can ameliorate these risks and clarify the scope of permissible opinions. Where the government calls a number of senior command representatives, trial counsel should assess 12 United States v. Eslinger, Nos. 10-0537/AR and 11-5002/AR which and how many are necessary to rebut the defense contention that the accused should be retained in the service. Third, “‘The Military Rules of Evidence [M.R.E.] are applicable to sentencing . . . thus providing procedural safeguards to ensure the reliability of evidence admitted during sentencing.’” United States v. Saferite, 59 M.J. 270, 273 (C.A.A.F. 2004) (alteration in original) (quoting United States v. McDonald, 55 M.J. 173, 176 (C.A.A.F.2001); Manual for Courts– Martial, United States, Analysis of the Rules for Courts-Martial app. 21 at A21-69 (2002 ed.)). Thus, a lay witness must always have a proper foundation to offer an opinion. See M.R.E. 701. In sum, although rebuttal testimony of the type in this case may raise some of the same concerns addressed by R.C.M. 1001(b)(5), that is different than concluding that this rule specifically applies to rebuttal evidence. We conclude that it does not. Rebuttal is governed by R.C.M. 1001(d), which does not contain the same restrictions as R.C.M. 1001(b)(5). Foundation for Opinions M.R.E. 701(a) requires that lay witness opinions or inferences be limited to those that are “rationally based on the perception of the witness.” In similar fashion, M.R.E. 602 provides that “[a] witness may not testify to a matter unless evidence is introduced sufficient to support a finding that the witness has personal knowledge of the matter.” See also Ohrt, 13 United States v. Eslinger, Nos. 10-0537/AR and 11-5002/AR 28 M.J. at 306 (discussing a parallel requirement found in R.C.M. 1001(b)(5)(B) that “the opinion envisioned by R.C.M. 1001(b)(5) can only be expressed by a witness who has a rational basis for his conclusions, founded upon the accused’s service performance and character”). As the Court of Criminal Appeals stated, “in the absence of such a foundation outlining these witnesses’ personal knowledge of appellant’s background or character, their subsequent testimony both lacks probity and increases the potential for prejudicial misuse of their opinions.” Eslinger, 2010 CCA LEXIS 64, at *34 (footnote omitted). Appellant argues that the Government’s witnesses did not have sufficient personal knowledge of the accused to proffer the opinions offered. We consider the testimony of each of the witnesses in issue. When asked the basis of his knowledge, MAJ Peltier agreed with the premise of the question that his opinion stemmed from what he “learned from Colonel Stoltz and the prosecutors in this case,” as well as “knowledge of the unit . . . [w]e’re very small.” MSG Stensgaard testified that he knew Appellant because “We did quite a bit of pre-mission training together prior to the OIF III deployment” and was aware of his two previous citations for drunk driving and criminal trespass conviction. COL Tovo testified that he knew who Appellant was and that he had a “good reputation as a soldier, particularly a combat 14 United States v. Eslinger, Nos. 10-0537/AR and 11-5002/AR soldier.” When asked the basis of his opinions, he stated “[Y]ou just listed four fairly significant instances of ill- discipline, and frankly, that’s more chances than we allow a guy.” SGM Krider stated that his opinion was based on the convictions for child pornography in this case, his “record of DUIs” and his criminal trespass conviction. The military judge sustained defense counsel’s objection to SGM Krider’s testimony and instructed the members to disregard it. Any infirmity with this testimony was cured by the military judge’s instruction and there is no indication in the record that the members did not follow the instruction. United States v. Taylor, 53 M.J. 195, 198 (C.A.A.F. 2000) (court members are presumed to follow the military judge’s instructions); United States v. Holt, 33 M.J. 400, 408 (C.M.A. 1991). CSM Sekelsky testified that he deployed with Appellant to Baghdad and “would see him occasionally at FOB 103.” He stated that he had occasional conversations with Appellant “[I would] just say, ‘Hi. How’re you doing,’ when I would see him. . . .” Although defense counsel objected to the testimony as cumulative, when asked if there was any other objection to the testimony, counsel replied “No, sir.” Thus any additional 15 United States v. Eslinger, Nos. 10-0537/AR and 11-5002/AR claimed infirmity regarding this testimony was affirmatively waived and is not subject to plain error review.4 In the case of MSG Stensgaard the foundational basis for the testimony was evident based on the structure of Special Forces units and the role of senior enlisted personnel within these commands. MSG Stensgaard was Appellant’s team sergeant for two years, trained with Appellant and deployed with him to Iraq. In short, this senior enlisted Special Forces soldier had a substantial personal foundation on which to render his opinions. The testimony of COL Tovo and MAJ Peltier present closer questions. According to MAJ Peltier’s testimony, his opinion of Appellant was based on what he learned from others as well as knowledge of the unit. Although not presenting the most compelling case, absent objection, it was not unreasonable for the military judge to infer that the executive officer of a Special Forces battalion would have direct and personal knowledge of a senior enlisted member in the command. 4 See United States v. Campos, 67 M.J. 330, 333 (C.A.A.F. 2008). However, we note that CSM Sekelsky’s opinion was based on his testimony that he was the command sergeant major of the 3d Battalion at the time of Appellant’s Operation Iraqi Freedom III deployment to Iraq. In this capacity, he would “see him occasionally” and of course was privy to the sort of information about senior noncommissioned officers that a command sergeant major is responsible for knowing. 16 United States v. Eslinger, Nos. 10-0537/AR and 11-5002/AR COL Tovo’s testimony was based on his standing as 10th Special Forces Group commander. According to COL Tovo the authorized medic strength for the Group was eighteen. His Group was below strength. In response to leading questions, COL Tovo indicated that his knowledge of Appellant was based on Appellant’s reputation in the command. He did not state that he had direct personal knowledge of Appellant. In our view, COL Tovo’s foundation in the record for expressing a personal opinion about Appellant was not as strong as it could have been. However, in the context of plain error review and in the context of the tightly knit and relatively small units that comprise the Army Special Forces community, in this case, the 10th Special Forces Group, we are not prepared to conclude, absent a record indicating otherwise, that the military judge abused his discretion in admitting the Group commander’s testimony. We agree with Appellant and the Court of Criminal Appeals that lay opinions must be derived from direct observation and judgment, but the military judge did not commit plain error by admitting the testimony of COL Tovo, MAJ Peltier, or MSG Stensgaard. Restated, it is not evident that there was a clear and obvious basis to exclude their testimony for lack of foundation. 17 United States v. Eslinger, Nos. 10-0537/AR and 11-5002/AR In any event, we need not ultimately determine if any of the admitted rebuttal testimony was obvious error, for even if so, any error was harmless. Prejudice Analysis Under the plain error test, after finding plain or obvious error we test for prejudice. That is, “We test the erroneous admission or exclusion of evidence during the sentencing portion of a court-martial to determine if the error substantially influenced the adjudged sentence.” Griggs, 61 M.J. at 410. While we do not find plain or obvious error in this case, for the sake of appellate thoroughness we consider the third prong of the plain error test, recognizing that opinions may reasonably vary as to whether an error was clear or obvious in the first instance. Where command influence concerns are raised, the application of prejudice analysis will also enhance confidence in the fairness of the system. Prejudice analysis is also useful because, as we recognized in Griggs, the line between an opinion on whether an accused should be returned to service or punitively discharged is a thin one. 61 M.J. at 409. Indeed the line can appear obscure if testimony that the accused is unworthy of continued service is viewed as a euphemism for a punitive discharge, as it no doubt often is. Intuitively, there is a certain fiction to the premise that the members will readily distinguish between an 18 United States v. Eslinger, Nos. 10-0537/AR and 11-5002/AR opinion on continued service and an opinion that the accused should be punitively discharged. But it is a distinction that we are confident that properly instructed members are capable of making. Here, the Government came closest to the line by asking each witness, “Do you want him in the Army?” But in the context of the defense witnesses stating their desire to have Appellant stay in the Army, this was not obvious error on rebuttal. As we weigh the factors in determining whether, if there was error, it was prejudicial, we weigh factors on both sides. On the one hand, Appellant’s combat service offered significant mitigation for members to consider on sentencing. He was a senior enlisted soldier with combat service as a medic. He had three combat tours and was awarded the Bronze Star with combat V. Finally, if the testimony in question was error, it came from senior officers in Appellant’s chain of command, which in theory makes their testimony more likely to influence the members. On the other hand, we find six factors that cut against a conclusion that Appellant was prejudiced. First, Appellant’s possession of child pornography was extensive. As noted by the lower court, Appellant collected it over time and in multiple locations. Eslinger, 2010 CCA LEXIS 64, at *3. It included over 1,700 images, including infants being sodomized and vaginally penetrated. Second, Appellant did not make the case 19 United States v. Eslinger, Nos. 10-0537/AR and 11-5002/AR that his conduct was in some manner the result of his combat experience. Third, Appellant faced a maximum punishment of thirty years of confinement, forfeiture of all pay and allowances, reduction to the lowest enlisted grade E-1, and a dishonorable discharge; he received three years of confinement, forfeiture of all pay and allowances, reduction to the lowest enlisted grade E-1, and a bad-conduct discharge. Fourth, Appellant had an extensive record of prior misconduct. This record included two General Officer Memoranda of Reprimand (GOMOR) for driving under the influence in 1999 and driving while intoxicated in 2004, and a stipulation of fact reflecting a civilian conviction in 2004 for third-degree criminal trespass. In addition, the military judge gave a standard instruction to the members to guide them in their decision on whether to award a punitive discharge and, if so, what kind. Finally, and relevant to all the other factors, Appellant was sentenced by a panel of six experienced members, including a colonel, two lieutenant colonels, a major, and two sergeants major. Considering these factors in the context of the test set out in United States v. Kerr, 51 M.J. 401, 405 (C.A.A.F. 1999), while we do not assume error, we agree with the lower court’s assessment: the possibility Appellant would have received less confinement or would have avoided a punitive discharge, absent the rebuttal testimony, was remote. We are confident that the 20 United States v. Eslinger, Nos. 10-0537/AR and 11-5002/AR testimony of COL Tovo, MAJ Peltier, or MSG Stensgaard did not substantially influence the members’ judgment on the sentence. CONCLUSION In view of the above, the decision of the United States Army Court of Criminal Appeals on the findings and the sentence is affirmed. 21 United States v. Eslinger, No. 10-0537/AR ERDMANN, Judge, with whom EFFRON, Chief Judge, joins (dissenting): This case involves four issues: (1) whether the Government established sufficient foundation for the testimony of the rebuttal witnesses it called during the sentencing portion of the court-martial; (2) whether that testimony was proper under United States v. Ohrt, 28 M.J. 301 (C.M.A. 1989); (3) whether the rebuttal testimony from five senior officers and senior noncommissioned officers from the command structure raised the specter of unlawful command influence; and (4) whether the admission of this testimony was prejudicial. As this court noted in United States v. Griggs, 61 M.J. 402, 409 (C.A.A.F. 2005): The chief concerns underlying these cases are the need to have “a rational basis for” an opinion concerning rehabilitation and the importance of avoiding command influence in the sentencing process. These concerns coincide with the UCMJ’s overarching concern regarding undue command influence. (Citations and quotation marks omitted.) I respectfully dissent from the majority’s conclusion that even if the admission of the rebuttal testimony constituted plain or obvious error, there was no prejudice. I agree with the United States Army Court of Criminal Appeals’ determination that the Government’s rebuttal testimony lacked proper foundation and its admission United States v. Eslinger, No. 10-0537/AR constituted plain and obvious error. However, I do not believe that the defense retention testimony, whether proper or not, opened the door to allow the Government to introduce otherwise inadmissible “euphemism” testimony on rebuttal. In addition, the military judge failed to give cautionary instructions addressing the proper purposes for which the rebuttal testimony was admitted and the potential for unlawful command influence. Under these circumstances, I cannot be confident that this improper testimony did not substantially influence the sentence. I would set aside the sentence and remand the case for a sentence rehearing. Relationship of R.C.M. 1001(b)(5) to R.C.M. 1001(d) I agree that there is no reference in Rule for Courts- Martial (R.C.M.) 1001 which would specifically make R.C.M. 1001(b)(5) applicable to R.C.M. 1001(d). However, the requirement for a witness to possess a rational basis for an opinion concerning the rehabilitative potential of an accused found in R.C.M. 1001(b)(5), is also embodied in both Military Rule of Evidence (M.R.E.) 602 and M.R.E. 701, and testimony admitted pursuant to R.C.M. 1001(d) must also satisfy those provisions.1 1 Rather than holding that R.C.M. 1001(b)(5) literally applied to R.C.M. 1001(d), the Court of Criminal Appeals appears to have looked to R.C.M. 1001(b) to inform its analysis as to whether the Government rebuttal witnesses 2 United States v. Eslinger, No. 10-0537/AR This requirement has been long recognized by this court: Mil.R.Evid. 701 governs admissibility of lay- opinion testimony, and it applies to evaluative statements offered under RCM 1001(b)(5). United States v. Susee, 25 MJ at 540. Only “opinions . . . which are . . . rationally based on the perception of the witness and . . . helpful to a clear understanding of the testimony of the witness o[r] the determination of a fact in issue” are admissible. Mil.R.Evid. 701. Thus, a foundation must be laid to demonstrate that the witness does possess sufficient information and knowledge about the accused -- his character, his performance of duty as a servicemember, his moral fiber, and his determination to be rehabilitated -- to give a “rationally based” opinion. . . . In United States v. Horner, . . . we tried to make it clear that “rehabilitative potential” refers to the accused. It is based upon an “assessment of . . . [the accused’s] character and potential.” 22 MJ at 296. Thus, a witness whose opinion is based upon factors other than an assessment of the accused’s service performance, character, and potential does not possess a rational basis for expressing an opinion. Ohrt, 28 M.J. at 303-04. Foundational Basis for Rebuttal Testimony The Government called five members of the 10th Special Forces Group command structure to rebut the defense sentencing witnesses: possessed the required rational basis for their expressed opinions, and then determined that they did not. United States v. Eslinger, No. ACM 20070335, slip op. at 17 (A. Ct. Crim. App. May 21, 2010). As the same foundational basis is required in M.R.E. 602 and M.R.E. 701, it was not error for the lower court to make that analogy. 3 United States v. Eslinger, No. 10-0537/AR Major Peltier Major (Maj.) Peltier was the Acting Battalion Commander of the 3rd Battalion, 10th Special Forces Group. He testified that it was his opinion, based on Eslinger’s pattern of misconduct, that he should not deploy with the unit or even be in the Army. On cross examination Maj. Peltier agreed that his opinion was based on what the Battalion Commander and prosecutors had told him. He did not testify as to any personal knowledge of Eslinger and acknowledged that he had never been on the same team with him and had never deployed with him. He testified that he had not been aware of Eslinger’s two General Officer Memorandums of Reprimand (GOMORs) for driving under the influence of alcohol or his civilian criminal conviction until the day before trial. Addressing the foundation established by the Government for Maj. Peltier’s testimony, the majority held that “it was not unreasonable for the military judge to infer that the executive officer of a Special Forces Battalion would have direct and personal knowledge of a senior enlisted member in the command.” I do not believe that such an inference meets the foundational standard of a “rationally based” opinion that is required for the admission of this evidence. See United States v. Kirk, 31 4 United States v. Eslinger, No. 10-0537/AR M.J. 84 (C.M.A. 1990).2 Maj. Peltier’s testimony was not rationally based and therefore lacked proper foundation. Sergeant Major Krider Sergeant Major (SGM) Krider was the Acting Command Sergeant Major for the 3rd Battalion, 10th Special Forces Group. He testified that he vaguely knew Eslinger, “in a distant manner.” He testified that Eslinger should not continue to serve in the Special Forces or the Army and primarily based that opinion on Eslinger’s conviction for possession of child pornography. The defense objected to the testimony of SGM Krider because his opinion was primarily based on Eslinger’s conviction for possession of child pornography and asked that the military judge instruct the members to disregard the testimony. In 2 In Kirk, 31 M.J. at 88, the court noted: In United States v. Horner, 22 MJ 294 (CMA 1986) this Court held that RCM 1001(b)(5), Manual for Courts-Martial, United States, 1984, requires that an admissible opinion on rehabilitative potential be based on an individual assessment of a servicemember’s character and potential. In United States v. Ohrt, 28 MJ 301, 304 (CMA 1989), this Court said that it must be shown that a commander expressing such an opinion “does possess sufficient information and knowledge about the accused -- his character, his performance of duty as a servicemember, his moral fiber, and his determination to be rehabilitated -- to give a “rationally based” opinion. 5 United States v. Eslinger, No. 10-0537/AR response, the military judge properly instructed the members to disregard the testimony of SGM Krider. Master Sergeant Stensgaard Master Sergeant (MSG) Stensgaard was a team sergeant in the tactical support detachment in the Group Support Battalion of the 10th Special Forces Group. MSG Stensgaard testified that he had been Eslinger’s team sergeant for two years and had trained and deployed with him. The majority held that sufficient foundation was established for MSG Stensgaard’s testimony and I agree. Colonel Tovo Colonel (Col.) Tovo was the Group Commander of the 10th Special Forces Group. He testified that he was aware that Eslinger had been convicted of possession of child pornography, had received two GOMORs for alcohol-related driving incidents, and had been convicted in civilian court of criminal trespass. As noted by the majority, Col. Tovo based his opinions on Eslinger’s reputation in the command and there is nothing in the record which indicates that he had any personal knowledge of Eslinger. He testified that he did not want Eslinger back in his unit, did not want to deploy with him, and did not want him in the Army. The majority recognized that the foundation for Col. Tovo’s testimony was “not as strong as it could have been” but in 6 United States v. Eslinger, No. 10-0537/AR the context of plain error held that the military judge did not abuse his discretion in admitting the testimony. As noted in United States v. Horner, this type of testimony is not helpful as “[t]he witnesses’ function in this area is to impart his/her special insight into the accused’s personal circumstances.” 22 M.J. 294, 296 (C.M.A. 1986). There is nothing in the record that establishes a rational basis for Col. Tovo’s insight into Eslinger’s personal circumstances or his potential for rehabilitation and therefore his testimony lacked a proper foundation. Command Sergeant Major Sekelsky Command Sergeant Major (CSM) Sekelsky was the 10th Special Forces Group Command Sergeant Major. The defense objected to CSM Sekelsky’s testimony as being cumulative with Col. Tovo’s testimony, but the military judge overruled the objection when he determined that CSM Sekelsky had more personal knowledge than Col. Tovo. The military judge then asked if the defense had any further objections to CSM Sekelsky’s testimony and defense counsel responded “No Sir.” I agree with the majority that this response affirmatively waived the issue of proper foundation for CSM Sekelsky’s testimony. 7 United States v. Eslinger, No. 10-0537/AR The rebuttal testimony of both Col. Tovo and Maj. Peltier lacked proper foundation. Given the numerous decisions of military courts on this issue, the military judge’s admission of that testimony constituted plain and obvious error. Inadmissible “Euphemism” Testimony on Rebuttal The majority recognized the “thin line between an opinion that an accused should be returned to duty” which is permissible testimony, and the “expression of an opinion regarding the appropriateness of a punitive discharge” which is impermissible testimony. In Griggs, 61 M.J. at 409, we explained: Obviously, an accused cannot return to serve in his unit if he receives a punitive discharge. But an explicit declaration that an accused should not receive a punitive discharge or that any such discharge should be of a certain severity is disallowed for the defense not because of R.C.M. 1001(b)(5)(D), but because such evidence invades the province of the members to decide alone on punishment. Ohrt, 28 M.J. at 305 (“The question of appropriateness of punishment is one which must be decided by the court-martial; it cannot be usurped by a witness.”). (Citation omitted.) This prohibition is not limited to express recommendations of a particular sentence, but also includes euphemisms: a commander as a sentencing witness cannot recommend a particular sentence to a court-martial 8 United States v. Eslinger, No. 10-0537/AR or employ euphemisms in his testimony which ineluctably lead to the same result. Here, the commander testified, “I think it would be, you know, a waste of Air Force resources to retain her.” This language rationally conveys the commander’s opinion that appellant should be separated, which is an impermissible comment under United States v. Ohrt. Kirk, 31 M.J. at 89 (citations omitted). This conclusion is based on the obvious fact that courts-martial have no authority to sentence an accused to any discharge other than a punitive discharge. When a senior officer or senior noncommissioned officer opines that an accused should not be in the Army (or other service), the message to the members is that the accused should receive a punitive discharge. Every Government rebuttal witness in this case testified that Eslinger should either not remain in the service or in the Army. This testimony infringed upon the province of the members and was improper.3 3 I do not agree with the Court of Criminal Appeals that because the defense presented retention evidence in the defense sentencing case that may have infringed upon the province of the members, that opened the door to allow the Government to introduce its own inadmissible testimony on rebuttal. Certainly the Government was entitled to present testimony as to whether others in the command wanted to continue to serve with Eslinger. However, even if we were to assume that the defense sentencing evidence infringed upon the province of the members, the Government should not be permitted to subsequently introduce inadmissible evidence to rebut the evidence presented by the defense, no matter how far the door has been opened. See also 9 United States v. Eslinger, No. 10-0537/AR As we recognized in Griggs, even when permissible testimony in this area is admitted, the military judge should provide “a tailored instruction focusing on the distinction between a punitive discharge, which is for the members to decide, and the willingness of a servicemember to serve with an accused again.” 61 M.J. at 409. No tailored instruction was provided in this case and the members were left with no guidance as to how to properly consider the rebuttal testimony. The Specter of Command Influence As recognized by the majority, when the Government presents rebuttal testimony to defense retention testimony, the military judge must be vigilant for the “specter of command influence.” This is particularly true when the rebuttal witnesses include the Group Commander, the Acting Battalion Commander, the Group Command Sergeant Major, and the Acting Battalion Command Sergeant Major. When senior officers and noncommissioned officers testify that an Eslinger, No. ACM 20070335, slip op. at 17. This particular theory of admissibility is sometimes called the “doctrine of curative admissibility” and notwithstanding this court’s passing references in United States v. Pompey, 33 M.J. 266, 270 n.2 (C.M.A. 1991), United States v. Banks, 36 M.J. 150, 164 n.15 (C.M.A. 1992), and United States v. Haimson, 17 C.M.A. 208, 224 n.2, 17 C.M.R. 208, 224 n.2 (1954), this court has not adopted the doctrine, nor should it. 10 United States v. Eslinger, No. 10-0537/AR accused should not be in the Army, not only does that testimony improperly invade the province of the panel, the specter of command influence is certainly present. We have often noted that: Congress and this court are concerned not only with eliminating actual unlawful command influence, but also with “eliminating even the appearance of unlawful command influence at courts-martial.” United States v. Rosser, 6 M.J. 267, 271 (C.M.A. 1979). United States v. Lewis, 63 M.J. 405, 415 (C.A.A.F. 2006). The majority correctly recognizes the import of a properly tailored instruction under these circumstances, but appears unconcerned that no tailored instruction was given in this case. The majority finds that the “standard instruction to the members to guide them in their decision on whether to award a punitive discharge” was sufficient. Eslinger, __ M.J. at __ (20). That instruction, however, did not address the issue of command influence. Where, as here, the Government rebuttal witnesses included the senior leadership of the 10th Special Forces Group and the battalion to which Eslinger was assigned, properly tailored instructions advising the members of the limited use for which the testimony was admitted and which also addressed the concerns of command influence, were essential. See Griggs, 61 M.J. at 409. 11 United States v. Eslinger, No. 10-0537/AR Under these circumstances, the lack of foundation for the testimony of the two senior rebuttal witnesses, the admission of impermissible euphemism testimony, and the lack of any tailored instructions constituted plain and obvious error. Prejudice “We test the erroneous admission or exclusion of evidence during the sentencing portion of a court-martial to determine if the error substantially influenced the adjudged sentence.” Id. at 410. The majority opinion relies on six factors in concluding that any error in regard to the admission of the rebuttal evidence was harmless. The majority opinion initially notes that “Appellant’s possession of child pornography was extensive. . . . Appellant collected it over time and in multiple locations [and i]t included 1,700 images, including infants being sodomized and vaginally penetrated.” Eslinger, __ M.J. at __ (19) (citation omitted). Notwithstanding the significance of these facts, the court-martial panel sentenced Eslinger to only a fraction of the maximum authorized confinement -- three out of a potential thirty years. Although I do not know the degree to which this sentence reflected the panel’s consideration of Eslinger’s years of service and the nature of that service, the 12 United States v. Eslinger, No. 10-0537/AR relative brief period of confinement demonstrates that the panel did not view the offenses as significantly diminishing the value of the evidence offered in extenuation and mitigation. As to the second factor, the majority observes that “Appellant did not make the case that his conduct was in some manner the result of his combat experience.” Id. at 19-20. It is not apparent why the members would have expected Eslinger to have made this argument, nor is it apparent why the absence of this argument has any bearing on the question of prejudice. Neither is it apparent how the absence of prejudice is demonstrated by the third factor relied on by the majority -- that “Appellant faced a maximum punishment of thirty years of confinement [and] a dishonorable discharge [and] he received three years of confinement [and] a bad-conduct discharge.” Id. at 20. As noted in response to the first factor, the vast disparity between the maximum punishment and the actual punishment reflects the willingness of the members to give substantial consideration to the specific facts and circumstances of the offenses and the offender. At the same time, the punishment of three years of confinement and a bad-conduct discharge is sufficiently consequential to demonstrate the potential prejudice from 13 United States v. Eslinger, No. 10-0537/AR the improper admission of sentencing evidence and lack of proper instructions as to how the panel should consider the rebuttal testimony and the potential for unlawful command influence. As to the fourth factor, the majority states that Eslinger “had an extensive record of prior misconduct.” Id. at 20. In fact, the record reflects three incidents in the course of Eslinger’s eighteen-year military career: two GOMORs for driving under the influence of alcohol; and a state court conviction for third degree criminal trespass, to which Eslinger pleaded no contest in 2004. None of these incidents, either individually or cumulatively, resulted in any action for an administrative separation. Eslinger continued to serve in the Army, including completion of his third combat tour in Iraq. While it was certainly appropriate for the court-martial panel to consider these incidents during sentencing, they did not constitute “an extensive record of prior misconduct” and did not make it inevitable that he would receive a punitive discharge at trial. The fifth factor relied upon by the majority is that “the military judge gave a standard instruction to the members to guide them in their decision on whether to award a punitive discharge and, if so, what kind.” Id. As noted 14 United States v. Eslinger, No. 10-0537/AR earlier, this instruction did not inform the members as to how they should consider the rebuttal testimony when evaluating the appropriateness of a punitive discharge, nor did it address the issue of potential command influence. Finally, the sixth factor relied upon by the majority notes that “Appellant was sentenced by a panel of six experienced members, including a colonel, two lieutenant colonels, a major, and two sergeants major.” Id. The record, however, does not indicate anything unusual with respect to this panel in terms of the application of the panel selection criteria under Article 25, UCMJ, 10 U.S.C. § 825 (requiring appointment of members “best qualified for the duty by reason of age, education, training, experience, length of service, and judicial temperament”), amount of prior court-martial experience, or likelihood to have more or less concern for the opinions of the command leadership. Summary The rebuttal testimony of Col. Tovo and Maj. Peltier lacked a proper foundation. The testimony of all the Government’s rebuttal witnesses included impermissible “euphemism” testimony that invaded the province of the members. These errors were compounded by the failure of the military judge to give essential cautionary instructions addressing the proper purposes of the 15 United States v. Eslinger, No. 10-0537/AR testimony and the potential for unlawful command influence. Under the circumstances, I cannot be confident that the improper testimony did not substantially influence the sentence. The sentence should be set aside and the case remanded for a new sentence hearing. 16