Bartholomew M. BERRY, Sergeant

U.S. Army, Appellant


No. 04-0240


Crim. App. No. 20000960


United States Court of Appeals for the Armed Forces


Argued October 20, 2004


Decided May 10, 2005


ERDMANN, J., delivered the opinion of the court, in which GIERKE, C.J., and EFFRON and BAKER, JJ., joined.  CRAWFORD, J., filed a separate opinion concurring in the result.




For Appellant:  Captain Charles A. Kuhfahl Jr. (argued); Colonel Robert D. Teetsel, Lieutenant Colonel Mark Tellitocci, and Major Sean S. Park (on brief); Colonel Mark Cremin.


For Appellee:  Captain Michael D. Wallace (argued); Colonel Steven T. Salata and Major Natalie A. Kolb (on brief).


Military Judge:  William T. Barto



This opinion is subject to editorial correction before final publication.



Judge ERDMANN delivered the opinion of the court.

     After entering a guilty plea to the lesser included offense of sodomy, Sergeant (SGT) Bartholomew M. Berry was tried by a general court-martial on a number of offenses.  Berry was convicted of the greater offense of committing sodomy by force and without consent, making a false official statement, committing an indecent act with another, and fraternization, in violation of Articles 125, 107, and 134, Uniform Code of Military Justice (UCMJ), 10 U.S.C. §§ 907, 925, 934 (2000), respectively.  Berry’s sentence, which was approved by the convening authority, included a bad-conduct discharge, confinement for twelve months, and forfeiture of all pay and allowances.

     The sole issue before this court involves the admission of evidence of a prior sexual assault under Military Rule of Evidence (M.R.E.) 413.  At trial, Berry’s defense to the charge of forcible sodomy was that Sergeant T (SGT T) had consented to the oral sex incident that was the basis for the charge.  To counter this defense the Government sought to introduce testimony from LS, who testified he had been the victim of a similar act by Berry eight years earlier.  The military judge found that the testimony was relevant and admissible under M.R.E. 413 and that ruling was affirmed by the United States Army Court of Criminal Appeals in an unpublished opinion.  United States v. Berry, ARMY 20000960 (A. Ct. Crim. App. Dec. 18, 2003). 

We granted review of the Army court’s decision to determine whether the military judge erred in admitting LS’s testimony under M.R.E. 413.  We find that although the testimony was relevant, the military judge erred in admitting it because he failed to do an adequate balancing test under M.R.E. 403 and that under a proper M.R.E. 403 balancing test, the testimony was inadmissible.  We also find the error to be prejudicial.      



     Berry’s conviction flows from events that occurred after a barbecue at his residence attended by Berry, SGT T, and Private First Class (PFC) H.  They drank a few beers at the barbecue and then went to a bar in town where they continued drinking.  After they had been at the bar a few hours, SGT T fell out of his chair while reaching for a cigarette and the three decided to return to Berry’s residence.  SGT T became sick when they arrived at Berry’s residence and he threw up several times.  The evidence at trial gave two different versions of the events that followed. 

In Berry’s sworn statement, in which he admitted to consensual sodomy with SGT T, Berry said that he helped SGT T clean himself up and helped him into Berry’s bed.  He then stated that he and SGT T began kissing and Berry “brought [SGT T’s] pants down, and that’s when [the oral sex] happened.”  SGT T did not tell Berry to stop, and Berry claimed that SGT T was awake at the time although “he was going in and out, he was pretty drunk.”  After the oral sex, Berry went to sleep and woke up later to drive SGT T to pick up his daughter from day care. 

     SGT T testified to a different version of events.  He stated that he felt very dizzy and very sick when they left the bar and that he threw up a number of times when they got back to Berry’s residence.  After lying on Berry’s bed to sleep it off, the next thing he remembered was looking up and seeing Berry on top of him and kissing him.  He testified that he tried to stop Berry, but that he could not move.  Berry then asked him if he “wanted to f[---],” and he said no.  Berry told SGT T that he had some condoms, and SGT T again said no.

The next thing he remembered was that Berry’s “head started moving down towards my genital area.”  Again SGT T “tried to move and tried to push him off, but I couldn’t . . . . I felt like I was paralyzed and I just could not move.”  The last thing he remembered was “a warm sensation on my genital area, but I am not sure because I don’t remember seeing him being on it . . .  I just remember him going down towards that area, and then a sensation there of feeling that he was putting his mouth on my genital area.”  SGT T also remembered someone trying to get his underwear back on him and that Berry then woke him up to pick up his daughter.

In addition to the two conflicting accounts of the evening,  LS testified at trial over the objections of Berry’s defense counsel.  LS was fifteen at the time of his testimony and he testified regarding an encounter he had with Berry eight years earlier when he was six and Berry was thirteen.  He and Berry both resided in military housing in Hawaii.  According to LS, one day they were playing with a group of children and Berry called him over behind a tree.  Berry pulled his own pants down, and “tried to get me to such [sic] on his penis.”  When LS said no, Berry reached over and pulled his pants down and “said he’d do it to me first.”  LS hesitated, but Berry convinced him and then proceeded to suck on his penis.  Berry then said it was his turn, and LS began to suck on Berry’s penis.  They were interrupted when “a guy ran out,” and both of them ran away.  On cross examination, LS admitted that there was no force used by Berry but that Berry talked him into participating.

The prosecution sought to have LS’s testimony admitted under M.R.E. 413, stating that “it is relevant to Sergeant Berry’s propensity to sexually assault those who are in a position of vulnerability.”  Trial counsel also stated that the evidence satisfied the M.R.E. 403 balancing test established by this court in United States v. Wright, 53 M.J. 476 (C.A.A.F. 2000), and therefore was not overly prejudicial. 

Berry’s defense counsel objected to the admission of this testimony on the basis that it was “extremely distracting, and. . . extremely prejudicial.”  The defense argued that LS’s memory of the event, on a scale of one to ten (with ten being very clear), was only about a six.  With regard to the probative value of the evidence, the defense counsel argued that it was childhood sexual experimentation and that it took place eight years prior to the event in question with no evidence that anything of the kind happened in between.  He further argued that there was no evidence of physical force or lack of consent.  The military judge found the testimony to be relevant and admissible under M.R.E. 413.

The military judge denied Berry’s motion to exclude LS’s testimony as follows:


The accused is charged with an offense of sexual assault . . . . The proffered evidence is evidence that the accused committed another act of or offense of sexual assault, and the evidence is relevant, under Military Rule of Evidence 401 and 402. The proffered evidence involves conduct that occurred over eight years ago.

The proffered evidence is similar to the charged misconduct because it involves taking advantage of a vulnerable victim.   [LS] was, approximately, 6[]years old at the time of the alleged sexual assault by the accused, who, despite his own youth, is considered by the court to have acted upon someone of tender years who was unable to consent at the time.

[SGT T] was also, apparently, vulnerable in that he was apparently unconscious or sleep [sic], or experiencing periods of partial consciousness, based on the government’s proffered [sic] at the time of the alleged assault by the accused.

Both offenses involve homosexual fellatio performed by the accused on another; and this proffered evidence involves only a single act, potentially, established by a single witness.  As such, the defense motion to exclude the testimony of [LS] is denied.


In both his opening and closing statements, trial counsel referred to Berry’s acts with LS and reminded the members that the encounter could be considered relevant “because [Berry] took advantage of a person in a vulnerable position just like he did here in the case that you’re deciding.”  The military judge’s instruction to the members with regard to LS’s testimony stated that it could be considered “for the purpose of its tendency, if any, to show that the accused has a propensity to commit nonconsensual sexual acts against unusually vulnerable persons.”

The Army court affirmed the ruling of the military judge that LS’s testimony was relevant under M.R.E. 401 and 402.  Although the military judge did not articulate an M.R.E. 403 balancing test on the record, the Army court found that the military judge had conducted a balancing test.  That conclusion was based on the factual findings made by the military judge and the arguments raised by the trial counsel regarding the factors that must be considered in an M.R.E. 403 balancing test.  Because it found that the military judge had conducted the required balancing test, the Army court reviewed his ruling for abuse of discretion and found that he had not abused his discretion in admitting the testimony of LS under M.R.E. 413.

On appeal before this court, Berry argues that LS’s testimony does not meet the threshold test for admissibility because it is not logically relevant.  Berry further argues that even if LS’s testimony is relevant, neither the military judge nor the Army court did a proper balancing test as required by M.R.E. 403 and by this court’s decision in Wright.  He suggests that had they done so they would have concluded that the testimony should not be admitted.  Berry concludes by arguing that he was prejudiced by the erroneous admission of this testimony.  In response, the Government argues that the testimony is relevant and the military judge conducted the required balancing test under M.R.E. 403 and properly concluded that LS’s testimony should be admitted.



     The granted issue1 asks whether the military judge erred in admitting evidence of uncharged sexual acts between Berry and LS that occurred eight years earlier when Berry was thirteen and LS was six.  This court has noted that inherent in M.R.E. 413 is a general presumption in favor of admission.  See Wright, 53 M.J. at 482-83.  However, we have also noted that it is a “constitutional requirement that evidence offered under Rule 413 be subjected to a thorough balancing test” under M.R.E. 403.  United States v. Dewrell, 55 M.J. 131, 138 (C.A.A.F. 2001).  Where that balancing test requires exclusion of the evidence, the presumption of admissibility is overcome.  See Wright, 53 M.J. at 482-83


M.R.E. 413 Threshold Requirements

This court has held that to admit evidence under M.R.E. 413, three threshold determinations must be made:

1.  The accused is charged with an offense of sexual assault – [M.R.E.] 413(a);

2.  “The evidence proffered is ‘evidence of the defendant’s commission of another offense of . . . sexual assault’”; and

3.  The evidence is relevant under [M.R.E.] 401 and 402.  United States v. Guardia, 135 F.3d 1326, 1328 (10th Cir. 1998).


Wright, 53 M.J. at 482

As the military judge noted, the first and second threshold requirements were met because Berry was charged with forcible sodomy without consent, which is an offense of sexual assault under the UCMJ, and Berry’s conduct with LS falls under that same definition.  See M.R.E. 413(d)(1) (defining “sexual assault”).  Moving to the third threshold requirement, we must determine whether the military judge abused his discretion in finding LS’s testimony relevant under M.R.E. 401.  See United States v. Gilbride, 56 M.J. 428, 430 (C.A.A.F. 2002) (citing United States v. Ayala, 43 M.J. 296 (C.A.A.F. 1995)).  A military judge abuses his discretion “if his application of the correct legal principles to the facts . . . is clearly unreasonable.”  United States v. Williams, 37 M.J. 352, 356 (C.M.A. 1993).

     Relevant evidence under M.R.E. 401 is “evidence having any tendency to make the existence of any fact that is of consequence to the determination of the action more probable or less probable than it would be without the evidence.”  The military judge found that “[t]he proffered evidence is similar to the charged misconduct because it involves taking advantage of a vulnerable victim.”  From strictly a propensity viewpoint, the evidence does show that Berry had participated in similar conduct in the past.  This evidence, therefore, does have some tendency to make it more probable that Berry committed a nonconsensual act against a vulnerable person and we conclude that the military judge did not abuse his discretion in finding the testimony to be relevant.  See Wright, 53 M.J. at 480 (noting legislative history shows that Federal Rule of Evidence (F.R.E.) 4132 creates an exception to M.R.E. 404(b)’s general prohibition against the use of a defendant’s propensity to commit crimes).


M.R.E. 403 Balancing Test

    Once the evidence meets these threshold requirements, a military judge must apply the balancing test of M.R.E. 403 under which the testimony may be excluded if its “probative value is substantially outweighed by the danger of unfair prejudice, confusion of the issues, or misleading the members.”  M.R.E. 403.  In conducting the M.R.E. 403 balancing test a military judge should consider the following factors:  the strength of the proof of the prior act; the probative weight of the evidence; the potential to present less prejudicial evidence; the possible distraction of the fact-finder; the time needed to prove the prior conduct; the temporal proximity of the prior event; the frequency of the acts; the presence of any intervening circumstances; and the relationship between the parties.  Wright, 53 M.J. at 482

     The military judge made minimal findings relating to the Wright factors and did not articulate any balancing of those factors on the record.  Where the military judge is required to do a balancing test under M.R.E. 403 and does not sufficiently articulate his balancing on the record, his evidentiary ruling will receive less deference from this court.  See Dewrell, 55 M.J. at 138 (citing United States v. Manns, 54 M.J. 164, 166 (C.A.A.F. 2000)); United States v. Bailey, 55 M.J. 38, 41 (C.A.A.F. 2000) (citing Manns, 54 M.J. at 166).

     The Wright elements addressed by the military judge tend to support admission of the testimony:  the proof of the prior similar act was strong because it involved the testimony of the only witness who had first-hand knowledge about the event; there was only one prior act which, as noted, could be established by a single witness and would not take an inordinate amount of time; and the act occurred eight years ago between acquaintances where the victim was in a vulnerable position.

     The military judge failed to address the relevant remaining Wright factors:3 the probative weight of the evidence, the frequency of the acts, the temporal proximity of the prior act and the presence of intervening circumstances, and the distraction of the fact-finder.

     While the military judge noted that the prior incident occurred eight years earlier where the victim was in a vulnerable position, he failed to analyze further the circumstances of the two incidents with regard to the probative weight of the previous incident.  The incident with LS occurred outside during the day between children, while the incident with SGT T occurred at night in private quarters between two adults after an evening of drinking.     

     The incident with LS occurred only once, eight years prior to Berry’s encounter with SGT T, when Berry was thirteen years old.  The length of time between the events alone is generally not enough to make a determination as to the admissibility of the testimony.  The circumstances surrounding the individual and the events that transpired in the intervening period must be taken into consideration.  Where a defendant was an adult at the time he committed the prior sexual assault, this court has found incidents occurring more than eight years prior to the charged incident to be relevant under M.R.E. 413.  See Dewrell, 55 M.J. at 137-38 (finding incidents occurring approximately seven to ten years earlier admissible); Bailey, 55 M.J. at 41 (finding incidents occurring three and one-half and ten years prior admissible).  A similar finding is not readily made where a prior incident is between children or adolescents.

     In United States v. McDonald, 59 M.J. 426 (C.A.A.F. 2004), we evaluated whether evidence of sexual acts that occurred twenty years prior to the charged act, when the defendant was thirteen, was erroneously admitted for the purpose of showing intent or a common plan under M.R.E. 404(b).  We concluded that the evidence was not relevant under that rule, and in so doing we noted the absence of “evidence of that 13-year-old adolescent’s mental and emotional state, sufficient to permit meaningful comparison with Appellant’s state of mind as an adult 20 years later.”  Id. at 430.

     During the eight years between the two incidents Berry grew from a child of thirteen to an adult of twenty-one.  There was no evidence introduced to show that Berry engaged in similar acts between the time he was thirteen and the time of the incident with SGT T eight years later.  Consistent with McDonald, we also note here that there is no evidence suggesting that Berry’s mens rea at twenty-one was the same as it was when he was a child of thirteen.  As one scholar has stated:

Between the ages of twelve and seventeen, adolescents undergo a critical period of transition during which they experience rapid transformations in emotional, intellectual, physical, and social capacities.  Even older adolescents, whose raw intellectual capacities may rival those of adults, have less experience on which to draw in making and evaluating choices.  In short, adolescents are not simply miniature adults.


Kim Taylor-Thompson, States of Mind/States if Development, 14 Stan. L. & Pol’y Rev. 143, 152-53 (2003) (footnotes omitted).  When projecting on a child the mens rea of an adult or extrapolating an adult mens rea from the acts of a child, military judges must take care to meaningfully analyze the different phases of the accused’s development rather than treat those phases as being unaffected by time, experience, and maturity.  Where a military judge finds that the prior “sexual assault” acts of a child or adolescent are probative to an act later committed as an adult, such a determination must be supported in the record by competent evidence.  Consequently the passage of eight years in this case constitutes a notable intervening circumstance between the two events at issue when coupled with Berry’s growth from childhood to adulthood during that time. 

     We then turn to another of the Wright factors not addressed by the military judge -- the possible distraction of the fact-finder that could result from admission of the testimony.  In Bailey we noted that the danger considered by this factor is that admission of the evidence may “result in a distracting mini-trial on a collateral issue.”  Bailey, 55 M.J. at 41.  

     The prosecution’s opening statement did not begin with a reference to the facts of the present case but rather with a reference to the earlier incident:  “[I]n 1992, the accused was a 13-year old living in Hawaii on a military installation.  He called a neighborhood child, a six-year old boy behind a tree, he pulled down the little boy’s pants and he convinced him to allow him to suc[k] on the child’s penis.”

     In his closing statement the prosecutor reminded the members that Berry “convinced [a] little boy to try to suck on his penis[]” even though “the little boy said, no.”  He noted that the encounter with LS could be considered relevant “because [Berry] took advantage of a person in a vulnerable position just like he did here in the case that you’re deciding.”

Under these circumstances it is evident that a “distracting mini-trial” on the collateral issue of the LS incident resulted from the admission of LS’s testimony and the prosecution’s pointed references.  The emphasis on “a neighborhood child,” “a little boy,” and “a six year-old boy” all characterized Berry in the eyes of the members as a child molester, one of the most unsympathetic characterizations that can be made.  

     Considering the already limited probative value of LS’s testimony, that value clearly was outweighed by the danger that the members were distracted from considering his testimony for its proper purpose.  Applying the appropriate deference to the ruling of the military judge, we find that LS’s testimony fails the M.R.E. 403 balancing test and that the military judge’s decision to admit LS’s testimony was in error.



“A finding or sentence of court-martial may not be held incorrect on the ground of an error of law unless the error materially prejudices the substantial rights of the accused.” Article 59(a), UCMJ, 10 U.S.C. § 859(a) (2000).  For a nonconstitutional error such as this one, the Government has the burden of demonstrating that “the error did not have a substantial influence on the findings.”  United States v. McCollum, 58 M.J. 323, 342 (C.A.A.F. 2003); see also United States v. Gunkle, 55 M.J. 26, 30 (C.A.A.F. 2001). 

In evaluating whether erroneous admission of Government evidence is harmless, this court uses a four-part test, weighing:  (1) the strength of the Government’s case, (2) the strength of the defense case, (3) the materiality of the evidence in question, and (4) the quality of the evidence in question.  See McDonald, 59 M.J. at 430 (citing United States v. Kerr, 51 M.J. 401, 405 (C.A.A.F. 1999)). 

     The central question with regard to this specification was whether SGT T had in fact consented to the sodomy or whether Berry had taken advantage of him when he was in a vulnerable drunken state.  The Government evidence on this issue consisted of expert testimony speculating that SGT T had been drugged and SGT T’s vague testimony regarding the incident.  The defense submitted no evidence, but set forth Berry’s version of events through cross-examination of the witnesses and his sworn statement in which he admitted to consensual sodomy.  The result was a “he said/he said” case, where SGT T’s credibility and the effectiveness of the defense cross-examination were all that the members had to rely upon. 

     LS’s testimony added the first-hand account by a fifteen-year-old boy that he was sodomized at the age of six by Berry.  Even though the testimony was admitted for the limited purpose of showing that Berry had a propensity to commit nonconsensual sexual acts against unusually vulnerable persons, due to the inflammatory nature of the testimony and the emphasis given the testimony by the Government, it was likely considered by the members as much more than propensity evidence.  Berry became not just a soldier who stood accused of forcible sodomy, but rather a child molester who was charged with the offense of forcible sodomy.  Based upon our review of the record, it appears that LS’s testimony improperly tipped the balance of the evidence and the Government has not met its burden of demonstrating that this improperly admitted evidence “did not have a substantial influence on the findings.”  McCollum, 58 M.J. at 342.

     Finding that Berry was prejudiced by the military judge’s error we turn to whether, if a rehearing on the affected findings is deemed impracticable, reassessment would be appropriate.  We find that, considering the inflammatory nature of the evidence to which the members were erroneously exposed, it would not be possible to “reliably determine what sentence would have been imposed at the trial level if the error had not occurred.”  United States v. Sales, 22 M.J. 305, 307 (C.M.A. 1986).  Reassessment, therefore, is not appropriate.  



     The decision of the United States Army Court of Criminal Appeals is reversed.  That part of the finding of the specification of Charge II stating “by force and without the consent of the said Sergeant [T]” and the sentence are set aside.  The remainder of the specification of Charge II and the remaining findings of guilty are affirmed.  The record of trial is returned to the Judge Advocate General of the Army.  A rehearing on the affected finding and sentence is authorized.  If a rehearing on the affected finding is deemed impracticable, a rehearing may be held on the sentence alone.






2 M.R.E. 413 was adopted from F.R.E. 413, and the analysis of M.R.E. 413 in the Manual for Courts-Martial, United States (2002 ed.)(MCM), references the legislative history of F.R.E. 413 regarding congressional intent for the rule.  MCM, Analysis of the Military Rules of Evidence A22-37.


3 We note that one of those factors not considered by the military judge is whether it would have been possible to admit, for the same purpose, evidence that was less prejudicial than LS’s testimony.  The defense argued on appeal before the Court of Criminal Appeals and before this court that a statement by Berry admitting to the encounter with LS was available for that purpose and should have been considered by the military judge.  However, the issue of whether less prejudicial evidence was available was not raised by the defense at trial and Berry’s statement was not admitted by the Court of Criminal Appeals.  We make no decision as to whether this statement would have been admissible at the trial level, we simply note that the statement is not part of the record on appeal.  We must evaluate the military judge’s ruling based on what was known to him at the time of trial.  Accordingly, we have denied Berry’s “Motion to Attach Defense Appellate Exhibit A,” filed with this court on October 18, 2004, and will not consider the substance of Berry’s statement concerning the prior event. 



CRAWFORD, Judge (concurring in the result).


I agree that the military judge abused his discretion by conducting a flawed analysis under Military Rule of Evidence (M.R.E.) 403, and would reverse on that ground.  However, I cannot join in the majority’s reasoning because the judge also abused his discretion by determining evidence of Appellant’s childhood act to be logically relevant under M.R.E. 401.

To be legally relevant, evidence must first be logically relevant.  If Appellant’s childhood sexual act is not legally relevant to show his propensity to commit similar acts as an adult because “there is no evidence suggesting that Berry’s mens rea at twenty-one was the same as it was when he was a child of thirteen,”  __ M.J. (15), then it was not logically relevant to show his propensity to commit similar acts in the first place.

Relevant evidence under M.R.E. 401 is “evidence having any tendency to make the existence of any fact that is of consequence to the determination of the action more probable or less probable than it would be without the evidence.”  In this case, the military judge found that “[t]he proffered evidence is similar to the charged misconduct because it involves taking advantage of a vulnerable victim.”  Without elaboration, the majority accepts that similarity of conduct as sufficient to establish logical relevance.  Given the facts of this case, however, and considering the purpose for which the evidence was offered, logical relevance requires more. 

 The Government offered Appellant’s childhood act under M.R.E. 413, which permits introduction of so-called “propensity evidence,” i.e., “[t]he rule permits the prosecution to use evidence of the accused's uncharged past sexual assaults for the purpose of demonstrating his propensity to commit the charged offenses.”  United States  v. Parker, 59 M.J. 195, 198 (C.A.A.F. 2003)(citation and internal quotation marks omitted).  The military judge denied the defense’s motion to exclude LS’s testimony and instructed the members that they may consider LS’s testimony “for the purpose of its tendency, if any, to show that the accused has a propensity to commit nonconsensual sexual acts against vulnerable persons.”  To be logically relevant to this purpose, the evidence must have some “tendency to make the existence of” appellant’s propensity “more probable . . . than it would be without the evidence.”  M.R.E. 401.  As we recognized in United States v. Wright, 53 M.J. 476 (C.A.A.F. 2000), M.R.E. 413 opens with a reminder that, while generally admissible, evidence must first be “relevant”: 

In a court-martial in which the accused is charged with an offense of sexual assault, evidence of the accused’s commission of one or more offenses of sexual assault is admissible, and may be considered for its bearing on any matter to which it is relevant.


M.R.E. 413(a)(emphasis added);1 Wright, 53 M.J. at 480.

Conceding that the military judge did not cite propensity as the “fact” to which this evidence was relevant, the majority then concludes that “[f]rom strictly a propensity viewpoint, the evidence does show that Berry had participated in similar conduct in the past.  This evidence, therefore, does have some tendency to make it more probable that Berry committed a nonconsensual act against a vulnerable person[.]” __ M.J. (10-11).  This statement is particularly troubling in light of the later conclusion that “there is no evidence suggesting that Berry’s mens rea at twenty-one was the same as it was when he was a child of thirteen.” __ M.J. (15).  From these statements, one must conclude that the mere happenstance of a similar, earlier act demonstrates per se relevance to propensity, even absent evidence or presumption of a similarity of mens rea,2 or, presumably, character.  Such a conclusion strongly suggests that even differences in mental competence (and certainly differences in cognitive and emotional development) are inapplicable to a threshold analysis under M.R.E. 401 -- a departure from the application of that rule that I cannot embrace, for it leads to the conclusion that the mindless act of an infant is per se logically relevant to prove the state of mind or character of that infant as an adult and would be admissible unless excluded for some other reason.  “[Evidence] may also be inadmissible as irrelevant because a link in the chain of facts is missing that is required to give probative value to the evidence.”  1 Wharton’s Criminal Evidence § 4-5 at 298-99 (footnote omitted); see also M.R.E. 104(b); c.f.  Johnson v. Elk Lake School District, 283 F.3d 138, 154-55 (3d Cir. 2002)(applying Fed. R. Evid. 104(b) to Fed. R. Evid. 413(d)).  In applying M.R.E. 413, I believe the majority has confused evidence relevant to state of mind and character3 with evidence relevant to happenstance, and in so doing, has departed from our recently announced analysis in United States v. McDonald, 59 M.J. 426 (C.A.A.F. 2004),4 rejecting evidence for its failure to satisfy M.R.E. 401.  In McDonald, this Court stated: 

The military judge found, and the court below agreed, that the evidence was logically relevant both as to “common plan” and “intent.”  We disagree.  Applying the second prong of Reynolds, we hold that the evidence of Appellant's uncharged acts was not logically relevant to show either a common plan or Appellant's intent.


59 M.J. at 429-30 (internal citation omitted).


Because M.R.E. 413, as employed in this case, demands logical relevance to a character trait –- propensity to commit sexual acts with vulnerable persons –– I would again conclude that evidence of sexual acts that occurred when the defendant was thirteen is not logically relevant to prove character or cognition of that child as an adult, absent “evidence of that 13-year-old adolescent’s mental and emotional state, sufficient to permit meaningful comparison with Appellant’s character as an adult.”  McDonald,  59 M.J. at 430.  While Appellant’s case presents a gap of only eight years between the acts, as opposed to the twenty years in McDonald, those intervening years share the same evidentiary deficiency:  failure to account for the effects of puberty and adolescence on either cognitive development or character.  In McDonald, we required the appellant to establish the logical relevance of a thirteen-year-old’s “mental and emotional state” to the state of mind of that child as an adult.  In this case, we examine the logical relevance of a thirteen-year-old’s acts to the “propensity” of that child as an adult to engage in similar acts.  I recognize that there is no meaningful distinction between the relevance we required in McDonald and the relevance we should require here.

Happenstance vs. character and state of mind.  In our practice, as in the federal district courts, the undertaking of an act is frequently relevant in and of itself, without regard to the state of mind of the actor in performing the act.  See, e.g., M.R.E. 304 (confessions and admissions), M.R.E. 321 (eyewitness identification), and M.R.E. 406 (habit and routine).  In the course of a trial, innumerable other contexts arise in which proof of an act is relevant merely to show that the act occurred (e.g., facts establishing nonvolitional elements of offenses).  In other contexts, relevance requires that happenstance be paired with a specific purpose (e.g., M.R.E. 608(b)(evidence of prior conduct must be probative of truthfulness or untruthfulness); M.R.E. 801(d)(1)(B)(prior consistent statement must precede motive to fabricate).

Character or state of mind evidence.  When intent, plan, purpose, or character are involved -- when the fact made more or less probable is a quality of cognition -- mere happenstance may not be sufficient to establish logical relevance.  McDonald, 59 M.J. at 430; United States v. Humpherys, 57 M.J. 83, 89-90 (C.A.A.F. 2002); United States v. Tanksley, 54 M.J. 169, 175 (C.A.A.F. 2000); United States v. Matthews, 53 M.J. 465, 473 (C.A.A.F. 2000).  M.R.E. 413 and 414 freed prosecutors from the restraints of M.R.E. 404(b) and for the first time permitted evidence of an accused’s past acts to demonstrate the propensity of the accused to commit other such acts.  Although “propensity” has been defined in slightly differing terms by various sources, the common theme is the focus on “nature” as the source of the inclination,5 reinforcing the classification of such evidence as “character evidence.”  In short, the evidence is offered to show that the character or nature of the accused is such that he or she is predisposed to commit the charged offense, based on similar conduct undertaken by the accused in the past.  As I see it, in the context of any comparative “state of mind” or character evidence, there is a roughly graduated scale -- with mere unity of identity at one end and absolute identity of cognitive state at the other -- along which such evidence will generally fall.  While I agree that mere unity of identity may be sufficient to establish relevance when reasonable identity of cognitive state can be assumed (as in the acts of most adults), I do not believe we should apply that assumption to bridge the frequently vast chasm of puberty and adolescence.  With this in mind, I believe our logic in McDonald must control our examination of this even more powerful evidence.

  As the majority notes, children are not miniature adults.  Evidence that an accused possessed some cognitive characteristic at age twenty-one might logically be presumed relevant to whether the accused had that same characteristic at age twenty-nine; however, consistent with our logic in McDonald, I cannot agree that evidence of a cognitive characteristic at age thirteen can be assumed to be logically relevant to whether the accused had that same cognitive characteristic at age twenty-one. 

For most teens, [risky or antisocial] behaviors are fleeting; they cease with maturity as individual identity becomes settled. Only a relatively small proportion of adolescents who experiment in risky or illegal activities develop entrenched patterns of problem behavior that persist into adulthood.


Laurence Steinberg & Elizabeth Scott, Less Guilty by Reason of Adolescence:  Developmental Immaturity, Diminished Responsibility, and the Juvenile Death Penalty, 58 American Psychologist 1009, 1014 (2003).  Professors Steinberg and Scott also note that:

of criminal careers indicate that the vast majority of adolescents who engage in criminal or delinquent behavior desist from crime as they mature into adulthood (Farrington, 1986).  Thus the criminal choices of typical young offenders differ from those of adults not only because the choice, qua choice, is deficient as the product of immature judgment, but also because the adolescent’s criminal act does not express the actor’s bad character.


Id. at 1015.


  In addition to the concerns we expressed in McDonald, we have also recognized that, even absent the complicating factors of puberty and adolescence, “[e]ven an individual with certain characteristics may have internal self-monitoring which may or may not cause them to act similarly in various situations.”  United States v. Dimberio, 56 M.J. 20, 27 (C.A.A.F. 2001).

  While I agree that M.R.E. 413 has significantly reduced the analytical importance of temporal proximity between the charged acts and prior, similar acts introduced under that rule, I cannot agree that the rule creates a bypass around M.R.E. 401 and 402 or creates a “happenstance equals relevance” equation.  Discussing Fed. R. Evid. 413, which uses language similar to that of M.R.E. 413, the Eighth Circuit noted that:  “We have previously stated that this rule supersedes Rule 404's prohibition against character evidence, allowing testimony of prior bad acts in sexual assault cases, provided that it is relevant.”  United States v. Bird, 372 F.3d 989, 992 (8th Cir. 2004).  In sexual assault and child molestation cases, evidence that the defendant committed a prior similar offense “may be considered for its bearing on any matter to which it is relevant,” including the defendant’s propensity to commit such offenses.  Fed. R. Evid. 413(a), 414(a).  “If relevant, such evidence is admissible unless its probative value is ‘substantially outweighed’ by one or more of the factors enumerated in Rule 403, including ‘the danger of unfair prejudice.’”  United States v. Gabe, 237 F.3d 954, 959 (8th Cir. 2001)(quoting United States v. LeCompte, 131 F.3d, 769 (8th Cir. 1997).

This is not to say here, nor did we so hold in McDonald, that the acts of a child cannot be relevant to determination of the state of mind of that child as an adult.  59 M.J. at 430.  Rather, in this arena of great potential probity and great potential prejudice, we must follow the rule and require that threshold relevance to the specified “fact” be demonstrated with the same level of scrutiny we would apply to any other evidence offered for any other purpose.  Expressed another way, we should not be so confident in the crucible of M.R.E. 403 that we assume, for purposes of M.R.E. 413, that happenstance equals relevance.  This is particularly so, given that in Wright we emphasized the importance of M.R.E. 401 and 402, and in McDonald we denounced the assumption that happenstance equals relevance with regard to similar evidence offered under M.R.E. 404(b).  Because I am not convinced that the military judge made that critical comparison in this case, I cannot express confidence in the lower court’s conclusion that there was no abuse of discretion in this quarter. 




1 We noted in Wright that M.R.E. 413 and Fed. R. Evid. 413 “are virtually the same.”  Wright, 53 M.J. at 480 n.4Accordingly, analysis of Fed. R. Evid. 413 also illuminates M.R.E. 413.  In a discussion of Fed. R. Evid. 413, two scholars commented:


The evidence permitted is broadly defined as any act that would be either a state or federal crime related to either sexual assault or child molestation.  The limitations appear to be only relevance and the requirement that the prosecution provide fifteen days notice of its intent to use evidence pursuant to these rules.” 1 Barbara E. Bergman and Nancy Hollander, Wharton’s Criminal Evidence (15th ed. 1997)(footnotes omitted)(commenting on the nearly identical text of Fed. R. Evid. 413 and 414).


§ 4.42 at 458-59.


2 Of course, mens rea and character are not congruent concepts, but in the context of this evidence and the purpose for which it was offered, the two are logically indistinguishable.


3 “Propensity evidence” is a form of character evidence.  See 2 Stephen A. Salzburg et al., Federal Rules of Evidence Manual § 413.02[2] (8th ed. 2001).


4 Although McDonald addressed M.R.E. 404(b), the concept of legal and logical relevance runs through the military evidentiary rules, including M.R.E. 401, 402, 403, 404(b), 413, and 14, along with the other § IV rules.


5 Webster’s Third New International Dictionary of the English Language (1961)(“a natural inclination”).  See, e.g., Merriam Webster’s Collegiate Dictionary (11th ed. 2003)(defining propensity as “an often intense natural inclination or preference); Webster’s New Word College Dictionary (4th ed. 1999)(propensity is a natural inclination or tendency”).


Home Page |  Opinions & Digest  |  Daily Journal  |  Scheduled Hearings  |  Search Site