(dissenting):
Four weeks after his birth, Jarod Dimberio sustained severe trauma, including injuries to his nose, eyes, and brain. The medical personnel who treated Jarod and examined his records testified that Jarod’s injuries were consistent with “non-accidental trauma” and that Jarod had been shaken violently. The evidence introduced at trial indicated that only two people had access to Jarod and the opportunity to inflict such injuries during the pertinent time period — appellant and his wife, Nicole. The evidence called upon the members to decide which parent was the perpetrator.
The prosecution, during its case-in-chief, focused significant attention on the state of mind of the perpetrator of a “shaken baby” crime. The military judge excluded critical, relevant defense evidence which squarely joined issue with the prosecution’s evidence concerning the state of mind of the perpetrator. Under the circumstances of this case, that ruling denied appellant his constitutional right to present a defense. See Chambers v. Mississippi, 410 U.S. 284, 294, 93 S.Ct. 1038, 35 L.Ed.2d 297 (1973); accord Davis v. Alaska, 415 U.S. 308, 317, 94 S.Ct. 1105, 39 L.Ed.2d 347 (1974), and Washington v. Texas, 388 U.S. 14, 19, 87 S.Ct. 1920, 18 L.Ed.2d 1019 (1967). I respectfully dissent.
I. BACKGROUND
A. Consideration of the Expert’s Testimony at Trial
There was no eyewitness testimony or other direct evidence as to who caused Jarod’s injuries. The Government, relying on circumstantial evidence, sought to explain to the members why appellant, as a parent, would have inflicted such violent injuries on his newborn child.
The Government introduced explicit medical expert testimony of Dr. Wacksman that “the constellation of symptoms” seen in Jarod was “typical shaken baby syndrome.” Dr. Lonergan, another prosecution expert witness, similarly testified on direct examination that Jarod “was a violently shaken baby.” On cross-examination, Dr. Wacksman testified that shaken baby injuries typically result when the response of a care giver to stressors results in an “acute, abrupt, momentary loss of control.” The experts’ testimony was accompanied by prosecution evidence that on the night of Jarod’s injuries, appellant had been tired and under work-related stress. The prosecution asked the factfinders to infer the following from this evidence: Jarod suffered injuries that typically are inflicted by care givers who react acutely and abrupt*31ly with momentary loss of control under stress; appellant was a care giver who was tired and under stress at the time in question; therefore, appellant cracked under stress and inflicted Jarod’s injuries.1
In response, the defense sought to rebut the inferences suggested by the Government and to develop alternative inferences. In support of the inference that he was not the person who had lost his composure and shaken Jarod, appellant introduced evidence that he was calm under pressure and responded well to stress, even when tired and upon being awakened from sleep.
To complement the evidence that he was not the likely perpetrator, appellant attempted to demonstrate that the only other possible perpetrator — Nicole — was the more likely culprit. Defense counsel sought to introduce the expert testimony of Dr. Sharbo, a forensic psychiatrist whom the military judge earlier had appointed to assist the defense in reviewing Nicole’s medical records. According to defense counsel’s proffer, Dr. Sharbo would state his expert opinion that Nicole had certain mental disorders and traits that historically had led to impulsivity and instability and that, under these circumstances, she “would not be expec[t]ed to handle stressful situations well.” He would base the testimony on a personal interview with Nicole, review of her medical and psychological records and the report of the investigation under Article 32, UCMJ, 10 USC § 832, and consultation (with Nicole’s permission) with her previous treating psychiatrist.
Defense counsel proffered to the military judge that the evidence would reveal that Nicole had “suffered from major depression for over 6 years, as noted in the records, and that this major depression was manifested in many ways, primarily, as far a[s] we are concerned, in her ability to handle stressful situations and anger problems.” This dovetailed, in the defense view, with: (1) “evidence of a stressful situation for [Nicole] that evening”; (2) evidence that “she was drinking that night” and that “when she drinks alcohol she has a much greater inability to control her anger and stress”; and (3) testimony from the prosecution’s expert witnesses that “in a shaken baby syndrome it normally is related to the ability of the care giver to handle a stressful situation, either as exhibited by the baby or by external factors that that care giver is experiencing at that particular time that might not be related to the baby in particular.”
The military judge questioned defense counsel concerning the “nexus” between evidence that Nicole had difficulty in handling stress and the implication that the defense sought to make that she would respond to stress with violence, particularly violence toward her baby. Defense counsel answered that the proffered evidence “is the type of evidence that we think is going to link her psyche, if you will, her personality disorder, her major depressive disorder, her long history of anger and stress control related problems to that night and her ability, because of that, to handle any particular stressful situation that occurred.” The military judge did not accept this connection, commenting that, “unless you can provide the link to her being violent that night, her acting out that night, then there is not a nexus there and you have a bridge that you have got to connect up some way.” Defense counsel responded:
Your honor, if people had seen Mrs. Dimberio acting out that night toward her son ■with all due respect, I don’t think we would be here in this courtroom today. The fact of the matter is abuse of the child in these situations is normally not seen by a third party____ [I]n these particular types of cases the doctors will tell you that it all deals with stress and the ability to deal with stress. This is what causes the care giver to do this to the child, so, if we have a care giver, the mother, who had complete and equal access to that child that evening, *32who has a psychological disorder, an axis disorder that causes them, such as major depression, which we understand does not abate over time, that that is relevant to determining whether or not we are talking about the identity of the perpetrator, it is more likely that that care giver is the one that could not handle the stressful situation that caused the injuries to Jarod.
Defense counsel reinforced that view in later argument to the military judge:
[I]f he is going to come in here and tell the court that she has a major depressive disorder and that she has a personality disorder that exhibits itself in an inability or a decreasability to handle stressful situations then I think that that opinion is admissible to the court members, not, your honor, anything specifically about past instances or past stressful situations or past alcohol use except as it applies to the doctor forming his opinion that in the opinion format without going into specifics he then provides to the jury. We do know that at least she told one doctor that night that she was stressed and we do know that she was drinking alcohol. I think that is the link there in regards to her personality disorders or major depression, if any, that she certainly has had for 6 years as far as major depression. I understand that it has not disappeared at will and the issue of identity and motive in this case.
Shortly after this argument, the military judge deferred ruling on the matter until later in the trial.
The Government subsequently “move[d] under its existing motion in limine to exclude this testimony [of Dr. Sharbo] on the grounds that it doesn’t meet the relevance criteria under [Mil.R.Evid.] 401 or 403.” The prosecution argued that the proposed defense testimony was not relevant because there was no evidence that Nicole had faced a stressful situation that evening or that she would respond to stress with violence toward children.
The military judge ruled that the testimony of Dr. Sharbo was inadmissible. Although he did not cite a specific basis for his opinion, he indicated that the evidence was not relevant because the defense did not demonstrate that Nicole had reacted violently in response to stress in the past. See Mil.R.Evid. 401 and 402.
B. Consideration of the Expert’s Testimony on Appeal
In the Court of Criminal Appeals, appellant renewed his argument that Dr. Sharbo’s testimony was relevant and admissible. 52 MJ 550 (1999). The court concluded that “while the military judge’s reliance on the fact that there was no evidence that Mrs. Dimberio had ever assaulted an infant may have been overly narrow in scope, this does not detract from the essential correctness of his ruling.” Id. at 557.
According to the Court of Criminal Appeals, “[t]he critical deficiency of appellant’s proffer was its reliance on a predisposition or profile, which in turn depended upon a trait of Mrs. Dimberio’s character. This is specifically prohibited by Mil.R.Evid. 404(a)____” Id. at 558. The court stated that it did “not read the rules of evidence or case law as permitting a trial to be decided by traits associated with a personality disorder,” and it concluded that “the military judge was well within his discretion in reaching [the] conclusion” that “Mrs. Dimberio’s personality disorder was not relevant and, therefore, not admissible.” Id. at 559.
In the present appeal, the granted issue requires our Court to determine whether the expert testimony was relevant and admissible. The majority would affirm the exclusion of Dr. Sharbo’s testimony on several bases. First, the majority asserts that appellant “did not offer an appropriate foundation for the introduction of ... opinion-type evidence” as to a character trait of Nicole. According to the majority, “The expert in this case had not known Mrs. Dimberio long enough to have formed a traditional opinion as to her character____” and “the defense [did not] offer specific instances of conduct by Mrs. Dimberio.” 56 MJ at 26. Second, the majority apparently concludes that the defense proffer did not establish the relevance of the evidence, noting that “what is missing here is an adequate proffer that this evidence of Mrs. Dimberio’s mental health *33problems had a nexus or link to behavioral traits of acting out and violence.” Id. at 27. Third, the majority holds that even if relevant, “the evidence would still be inadmissible under [Mil.R.Evid.] 403’s balancing test” because the proffered evidence was “both speculative and potentially confusing to the members,” absent a showing that Nicole’s problems were tied to violence. Id. at 27. Finally, the majority indicates that any error was harmless because “appellant did present his defense to the court members through other means,” including evidence that Nicole was under stress at the time in question and that she apparently had been drinking alcohol. Id. at 27.
II. ADMISSIBILITY OF THE EXPERT’S TESTIMONY
A. Foundation for Expert Opinion
Mil.R.Evid. 405(a) provides that, whenever “evidence of character or a trait of character of a person is admissible, -proof may be made ... by testimony in the form of an opinion.”2 This rule includes testimony of the opinion of a psychiatrist that is based on a professionally satisfactory foundation. See Advisory Committee’s Note to Fed.R.Evid. 405, In re Townsend, 56 F. 222 (1893) (acceptable form of opinion testimony includes “the opinion of the psychiatrist based upon examination and testing”); United States v. St. Jean, 45 MJ 435, 442-44 (1996) (in murder trial where victim’s state of mind was in issue, testimony of a psychiatrist, who had examined copious records and documents relating to the victim and the crime, was admissible to offer his opinion that he detected no indication that the victim was either depressed or “highly impulsive,” which he already had testified were characteristics of persons with a high risk of committing suicide); State v. Christensen, 129 Ariz. 32, 628 P.2d 580, 582 (1981) (in murder trial where the defendant’s premeditation was in issue, testimony of a psychiatrist, who had interviewed the defendant and had reviewed tests administered to him, was admissible to offer opinion that the defendant “had difficulty dealing with stress and in stressful situations his actions were more reflexive than reflective”). See also 3 Jones on Evidence: Civil and Criminal § 16:24 at 152-54 (7th ed.1998); Stephen A. Saltzburg, Michael M. Martin, & Daniel J. Capra, 1 Federal Rules of Evidence Manual (hereafter Saltzburg) 402-03 (7th ed.1998), discussing United States v. Staggs, 553 F.2d 1073 (7th Cir.1977) (“In dictum, the Court seemed to approve the use of an expert opinion as to the character of a criminal defendant, which Rules 404(a) and 405 do not prohibit but which generally was not permissible at common law.”). See also Saltzburg, supra at 402, discussing United States v. Roberts, 887 F.2d 534 (5th Cir.1989) (“The Court held it was error, but harmless, to exclude testimony of a psychologist that the personality of a defendant charged with cocaine offenses was consistent with his claimed activity as a self-appointed vigilante.”).
In the present case, Dr. Sharbo personally interviewed Nicole, discussed the matter with her previous treating psychiatrist, and reviewed Nicole’s medical and psychological records as well as the Article 32 investigation report. These actions established a legally sufficient foundation for Dr. Sharbo to offer the expert psychiatric opinion testimony proffered by the defense. See Mil.R.Evid. 703, Bases of Opinion Testimony by Experts; Stephen A. Saltzburg, Lee D. Sehinasi, and David A. Schlueter, Military Rules of Evidence Manual 865 (4th ed.1997) (“Under Rule 703, an expert may base her opinion upon facts or data that she has perceived, learned from study or experiment, or been told about, either by watching the proceeding *34in court, or from other sources outside court.”); id. at 841 (“[P]ursuant to Rule 703, counsel may rely on expert witnesses to provide opinion testimony which is not based on first hand knowledge or observation.”); State v. Christensen, supra (in murder trial, psychiatric evidence as to defendant’s proclivity toward reflexive actions under stress was admissible on question whether appellant killed his wife with premeditation or impulsively). Cf. St. Jean, supra.at 444 (“[V]icarious fact-gathering is expressly permissible and normal in the medical, psychiatric, and psychological fields.”).
B. Relevance of the Expert’s Testimony
The prosecution’s theory of the case and the evidence introduced by the prosecution placed the psychological condition of the perpetrator squarely at the center of this trial. Only one of two people could have shaken Jarod and caused these injuries— appellant or Nicole. There was no evidence that either appellant or Nicole previously had responded to stress with violence in general or violence against a child in particular. On this occasion, however, one of them did so. The heart of appellant’s defense sought to focus the members on the question of whether it was appellant or Nicole who had responded to stress with violence against them child.
In this context, the prosecution emphasized factors relevant to appellant’s mental condition — that he had been tired and under stress on the night in question. The prosecution — consistent with their expert’s opinion — did not attempt to show that he had reacted to stress with violence in the past or that the stress preceding the incident was necessarily related to the baby. The proffered expert testimony as to the impact of Nicole’s psychological condition on her ability to cope with stress was at least as compelling as the prosecution’s evidence concerning appellant.
“[Ajnything that can help rationally decide disputed issues and be helpful to the finder of fact is relevant____ If evidence is of any value at all, it qualifies under the Rule.” Military Rules of Evidence Manual, supra at 474. Where the prosecution’s own evidence indicated that the perpetrator probably was someone who cracked under stress without necessarily having done so in the past — and without any showing that the stress was caused by the baby — the military judge erred in concluding that it was “irrelevant” that Nicole’s mental condition was such that she could be expected to have difficulty with stress. Under these circumstances, the excluded testimony of Dr. Sharbo would have had some “tendency to make the existence of any fact that is of consequence to the determination of the action more probable ... than it would be without the evidence.” Mil. R.Evid. 401.
C. MilR.Evid. J/.03
Mil.R.Evid. 403 precludes admission of relevant evidence under six circumstances. Even though “relevant, evidence may be excluded if its probative value is substantially outweighed by the danger of unfair prejudice, confusion of the issues, or misleading the members, or by considerations of undue delay, waste of time, or needless presentation of cumulative evidence.”
The military judge commented at one point that without some showing that Nicole’s mental condition would lead her to respond to stress with violence, the evidence in question “should be excluded under any type of [MiLR.Evid.] 403 balancing test.” The military judge, however, did not indicate what was so objectionable about the evidence that it substantially outweighed the probative value. In the absence of any analysis by the military judge, the majority attempts to perform the balancing test under Mil.R.Evid. 403, asserting that the proffered evidence would have been “both speculative and potentially confusing to the members.” 56 MJ at 27. The majority offers no analysis, however, as to why this evidence would have been so flawed.
Moreover, whether it would have been “potentially confusing” is not the test under Mil.R.Evid. 403; rather, the test is whether such difficulties as danger of unfair prejudice or confusion of the issues “substantially out-weight]” the “probative value” of the evidence. Mil.R.Evid. 403 “is constructed to favor the admission of evidence, as a result *35exclusion of otherwise relevant testimony should be rarely invoked.” Military Rules of Evidence Manual, supra at 490; see United States v. Roberts, 88 F.3d 872 (10th Cir. 1996); United States v. Mende, 43 F.3d 1298 (9th Cir.1995); United States v. Terzado-Madruga, 897 F.2d 1099 (11th Cir.1990). “The use of the word ‘substantially’ in the Rule suggests that in close cases the drafters intended that evidence should be admitted rather than excluded.” Military Rules of Evidence Manual, supra at 491. See United States v. Mende, supra; United States v. Krenzelok, 874 F.2d 480 (7th Cir.1989).
Furthermore, the “prejudice” that the rule seeks to avoid is the danger of “unfair prejudice” — that is, the danger that the evidence will be used “for something other than its logical, probative force.” Military Rules of Evidence Manual, supra at 492.3 The nature of the proffered evidence — given the context of this trial and the prosecution’s evidence — would have sharpened the issues and would have provided a complete picture of the defense theory for the members’ consideration without being unfairly prejudicial to the Government, confusing the issues, or misleading the members. Under these circumstances, Mil.R.Evid. 403 does not provide a basis for exclusion of the expert’s testimony.
D. MilREvid. kOh
Mil.R.Evid. 404(a) generally excludes “[ejvidence of a person’s character or a trait of a person’s chai'acter ... for the purpose of proving that the person acted in conformity therewith on a particular occasion____” The rule is based on the premise that character evidence offers “little probative value” in such circumstances, while creating a significant risk of diverting the members’ deliberations through the interjection of irrelevant issues. See Military Rules of Evidence Manual, supra at 524.
Subsections (l)-(3) of Mil.R.Evid. 404(a) recognize three exceptions to the general rule of inadmissibility: (1) evidence of a pertinent character trait of an accused offered by the accused, or by the prosecution to rebut it; (2) evidence of a pertinent character trait of the victim offered by the accused, or by the prosecution to rebut it; and (3) evidence of the character of a witness offered to impeach the witness under Mil.R.Evid. 607-609.
One explanation for these historical exceptions is that when an accused initiates use of a character trait “to exonerate himself, the problem of prejudice is altogether different. Now, knowledge of the accused’s character may prejudice the jury in his favor, but the magnitude of the prejudice or its social cost is thought to be less.” 1 McCormick on Evidence § 191 at 673 (5th ed.1999) (emphasis in original). Cf. 1 Federal Rules of Evidence Manual, supra at 374 (the rationale for the exception permitting an accused to introduce evidence of his own character “is that the defendant deserves the benefit of all reasonable doubts and that good character may produce a reasonable doubt.”). Similarly, in commenting on the exception relating to the defense introduction of evidence of the victim’s character, the McCormick text states:
That the character of the victim is being proved renders inapposite the usual concern over the untoward impact of evidence of the defendant’s poor character on the jury’s assessment of the case against the defendant. There is, however, a risk of a different form of prejudice. Learning of the victim’s bad character could lead the jury to think that the victim merely “got what he deserved” and to acquit for that reason. Nevertheless, at least in murder and perhaps in battery cases as well, when the identity of the first aggressor is really in doubt, the probative value of the evidence ordinarily justifies taking this risk.
*36McCormick, supra at 681 (emphasis in original).
Mil.R.Evid. 404(a) does not express an exception permitting the defense to introduce evidence of a relevant character trait of a third-party alternate perpetrator. Nonetheless, the considerations applicable to both the general rule of exclusion and the specific exceptions would seem to apply with equal force to an exception in this area. Tiller’s 1983 revised edition of 1A Wigmore, Evidence § 68 at 1444 contains the following observation:
[I]f one takes the view that character evidence is relevant and that the main reason for its usual exclusion is the danger of prejudice to the parties, it is not insensible to take the view that evidence of the character of third persons should be admissible when there is no substantial danger that the trier of fact will draw inferences about the character of the parties as a result of his views of the character of the nonparties.
(Citations omitted.) The authors quote the following passage from Professor Wigmore’s third edition of the Treatise that pointedly reflects this view:
Where the character offered is that of a third person, not a party to the cause, the reasons of policy for exclusion seem to disappear or become inconsiderable; hence, if there is any relevancy in the fact of character, i.e., if some act is involved upon the probability of which a moral trait can throw light, the character may well be received.
(Citations omitted.)
In this case, the disputed evidence of Dr. Sharbo’s testimony was critically important to the defense and went directly to the heart of the case, as shaped by the prosecution, and the question before the members: Which of two parents had violently injured Jarod? In that context, the relevance of the evidence was great and the risk of unfair prejudice and confusion was minimal, consistent with the rationale underlying Mil.R.Evid. 404. Under those circumstances, the rule cannot impede appellant’s constitutional right to a fair trial and a full presentation of his defense theory that Nicole committed the crime. See Chambers v. Mississippi, 410 U.S. 284, 93 S.Ct. 1038, 35 L.Ed.2d 297 (1973); Cikora v. Dugger, 840 F.2d 893 (11th Cir.1988). Cf. DePetris v. Kuykendall, 239 F.3d 1057, 1062 (9th Cir.2001) (trial court’s erroneous exclusion of crucial defense evidence as irrelevant “went to the heart of the defense” and, so, denied defendant’s “Fifth Amendment due process right to a fair trial” and her “Sixth Amendment right to present a defense,” citing Chambers v. Mississippi, supra).
III. PREJUDICE
The majority takes comfort in the fact that “appellant did present his defense to the court members through other means,” 56 MJ at 27, noting that the members heard evidence that Nicole was under stress and “ ‘had the smell of alcohol about her.’ ” The majority concludes: “The only area which appellant’s defense counsel was not allowed to explore was Nicole Dimberio’s mental health diagnosis and its link to the baby.” Id. at 27.
Without evidence of Nicole’s mental health diagnosis and the expert’s testimony concerning the likelihood that she would have difficulty handling stress, particularly when under the influence of alcohol, the defense was deprived of the testimony necessary to explain the significance of the evidence that she was under stress and had been drinking. It is not unusual for persons to be under stress, to drink alcohol, or both. What was missing in this case was the opportunity for the defense to present evidence that gave meaning to these factors — that this particular person had a mental condition that could cause her to respond in a certain way when under stress and particularly when drinking alcohol. Under these circumstances, the exclusion of the proffered expert testimony of Dr. Sharbo materially prejudiced the substantial rights of appellant. See Art. 59(a), UCMJ, 10 USC § 859(a).
. Trial counsel’s rebuttal argument during closing relied on Dr. Wacksman’s testimony, contending that this “is most likely a shaken baby rather than an impact trauma____” In arguing that appellant had the mental state to inflict such injuries, trial counsel argued that appellant had been “tired” that night and had "become upset” with Jarod. He contended that the evidence showed a “dehberate shaking of the baby out of frustration, out of anger, out of being upset.”
. The commentators are divided as to whether opinion testimony like that proffered by the defense in this case implicates traits of character within the meaning of Mil.R.Evid. 404. Compare, e.g., 2 Weinstein’s Federal Evidence § 405.04[2][c] at 405-28 to 405-29; 3 Jones on Evidence: Civil and Criminal § 16:24 at 152-54 (7th ed.1998), with 1A Wigmore, Evidence § 83 at 1599 (Tillers rev.1983); Stephen A. Saltzburg, Michael M. Martin, and Daniel J. Capra, 1 Federal Rules of Evidence Manual 529, Advisory Committee’s Note on original Fed.R.Evid. 404(a) (7th ed.1998); and Christopher B. Mueller and Laird C. Kirkpatrick, 1 Federal Evidence § 101 at 522 (2d ed.1994). See also Wigmore, supra § 51 at 1145 — 46.
. "[P]ractice often demonstrates that counsel’s arguments for excluding testimony largely consist of undefined, conclusoiy references to the evidence’s prejudicial effects, rather than pointed demonstrations of the evidence’s unfairly prejudicial impact on the court members’ ability to properly evaluate the other admissible evidence and reach an appropriate, non-emotional, result thereon. Unless counsel can articulate why the evidence will be unfairly used, it can be admitted.” Id. at 492-93. (last emphasis added).