delivered the Opinion of the Court.
¶1 Kelly Dale Clark (Clark) appeals the judgment of conviction for sexual assault and the order denying his motion for a new trial entered in the Twentieth Judicial District Court, Lake County. We affirm in part, reverse in part, and remand for further proceedings.
¶2 We address the following issues on appeal:
¶3 Did the District Court commit plain error by not instructing the *10 jury that it must be unanimous as to the means by which Clark subjected his stepdaughter to sexual contact?
¶4 Did the District Court abuse its discretion by denying Clark’s motion for a new trial after the complaining witness recanted her testimony?
BACKGROUND
¶5 On or around July 18, 2002, Clark’s fourteen-year-old stepdaughter, T.C., came home from babysitting and changed into a bikini in preparation for mowing the lawn. Her mother was gone for the evening, working. Clark came home from golfing and laid on the couch in the living room. According to T.C.’s later testimony, as she came into the living room Clark asked her if one of her breasts was bigger than the other, and he directed her to kneel beside him, which she did. T.C. stated that Clark then placed his hands on her breasts, said that he could not tell if one was larger than the other, then pushed the bikini top up, exposing both breasts. T.C. covered herself with her hands, but Clark said, “It’s not like I haven’t seen boobs before.” T.C. testified that Clark then showed her “boob exercises” by pushing T.C.’s breasts up, down, and sideways. Afterward, Clark told T.C. that she did not need to tell her mom about what happened.
¶6 One week later, T.C. told her mother, Christina, what Clark had done, and Christina called the police. Having obtained a restraining order against Clark, Christina and T.C. returned home. That night, Clark came to the house to confront Christina about the incident. Clark and Christina argued, culminating in Christina retrieving a gun from her bedroom, Clark retreating outside, and T.C. calling 911. Later that night, Clark was arrested. Within a few weeks, Clark and Christina were back together.
¶7 In February 2003, Clark was tried for sexual assault and violating an order of protection. 1 At trial, T.C.’s credibility was an issue, and there was evidence that she had lied on several occasions in the past and that she had changed her story about what had happened with Clark in the living room. Christina testified at trial that, on occasion, she had encouraged T.C. to lie in order to avoid what she perceived to be overly harsh punishments from Clark.
¶8 Christina also testified that in the weeks and months after the incident she began to disbelieve T.C.’s version of events. In describing *11 at trial the confrontation she had with her daughter on the matter, Christina said the following:
I wanted to talk to her about it, and I told her that I didn’t feel that she was being truthful. I didn’t feel that she was telling the whole truth. I also felt that she was telling a lot of things that weren’t true.
And she told me that she was. And I told her, no, you aren’t; you’re not telling the truth. I don’t believe you. And I told her that when she wants to come and tell me the truth, then I’ll be there.
And I think it was-I don’t know how much time passed. She came back in the living room and she told me she was sorry that she had done this, and that Kelly didn’t do all of those things that she said he had done.
After this conversation, Christina instructed T.C. to call Detective Doyle, to whom they had originally reported the incident in late July. On the phone with the detective, T.C. recanted her story, saying that it did not happen as she had described and that the incident was not "perverted.”
¶9 On October 23, 2002, Detective Doyle and Deputy County Attorney Young interviewed T.C. in a room at her school while Christina waited in the hallway outside where should could not hear the conversation. In that interview, T.C. told a substantially similar story as she had in her first report to the police. At trial, the State had T.C. read from the transcript of that interview:
Q. [by the State]... I’m going to read the question and I want you to read me the answer back; okay?
A. [by T.C.] Yeah.
Q. Okay. “What made you call me up that day and tell me that it was not true.”
A. “Well, my mom had talked to me that day and, uh, we were talking about Kelly and what would happen to him if she-and she said she was going-she was getting sick and tired of listening to me telling all the-she thought I was lying and so she wanted me to tell the truth, so I lied because I didn’t want her to get mad. And she said that I couldn’t be in her life if I didn’t tell the truth. So I had no choice.”
Q. Your mom told you that she was getting sick and tired of listening to your lies and that you couldn’t be in her life if you kept it up?
A. Yes.
¶10 In her trial testimony, T.C. admitted that she had lied about some *12 of the details she had initially provided to the police about the incident with Clark. In addition, she stated that, after the prosecutor had explained what sexual assault was, she did not believe she had been sexually assaulted. She maintained that what Clark had done to her was not “perverted” or “sexual” but was done with the intention of helping her. T.C. also stated that she wanted to resume living with Clark.
¶11 Clark’s trial testimony differed from T.C.’s in several respects. He disputed T.C.’s contention that he initiated the conversation about her breasts. Clark asserted that T.C. mentioned that her breasts were starting to grow and then came over and knelt beside him by her own volition. Clark stated that T.C. asked him whether her breasts were even, and then, to his great surprise, she lifted her bikini top, exposing her breasts. He maintained that he reached out and touched her with two fingers on the inside of her breast and said that that one was larger than the other. After just a few seconds, T.C. put her top back on and started doing the breast exercises that Clark had previously described to her.
¶12 Both T.C. and Clark testified that when she had any questions about sexual matters, she would come to her stepfather rather than her mother, and that it was not unusual for T.C. and Clark to have these conversations.
¶13 Despite the attacks on T.C.’s credibility, T.C.’s testimony about her interpretation of the incident with Clark, and Clark’s testimony as to his version of events, the jury returned a verdict of guilty.
¶14 Soon after the trial, T.C. went to live with Alana Myers and her husband, a pastor of a church in Bigfork, though T.C. would also periodically spend a few days at a time with Christina. One evening not long after the trial, Myers and T.C. had a direct and open conversation in which Myers, cognizant of T.C.’s mendacious tendencies, encouraged T.C. to be completely open and honest about what had happened with Clark, though she was careful not to suggest an answer. T.C. asked what she could do, and Myers said that T.C. could write a letter to Clark’s attorney. Two to four days later, without reminder or prompting from Myers, T.C. showed her a rough draft of a letter in which T.C. partially recanted her trial testimony. After Myers helped T.C. clear up some grammatical and punctuation errors, the letter was given to Clark’s attorney.
¶15 In March 2003, Clark moved for a new trial asserting that T.C. had recanted her trial testimony and that the recantation was new evidence. At a hearing on the matter, T.C. explained that she had lied *13 on the stand because she was confused and scared. T.C. stated that she initiated the contact between her and Clark, that she knelt down without being asked by Clark, and that she, not Clark, lifted her bathing suit top. Though she testified that Clark did not grope her breasts in any way, T.C. maintained that Clark did touch her on the side of her breast and that she was surprised that he did so. Clark argued that this was newly discovered evidence which entitled him to a new trial. The District Court denied the motion.
¶16 The District Court sentenced Clark to twenty years imprisonment, sixteen suspended, for the sexual assault conviction and six months imprisonment, all suspended, for violating an order of protection. Clark appeals.
STANDARD OF REVIEW
¶17 This Court explained the standards for applying plain error review in
State v. Finley
(1996),
[T]his Court may discretionarily review claimed errors that implicate a criminal defendants fundamental constitutional rights, even if no contemporaneous objection is made ... where failing to review the claimed error at issue may result in a manifest miscarriage of justice, may leave unsettled the question of the fundamental fairness of the trial or proceedings, or may compromise the integrity of the judicial process .... [W]e will henceforth use our inherent power of common law plain error review sparingly, on a casebycase basis ....
¶18 Generally, “[t]he standard of review of a district court’s ruling on a motion for new trial is whether the district court abused its discretion.”
State v. McCarthy,
DISCUSSION
¶19 Did the District Court commit plain error by not instructing the jury that it must be unanimous as to the means by which Clark subjected his stepdaughter to sexual contact?
¶20 The jury convicted Clark of sexual assault as defined in § 455502(1), MCA (2001): “A person who knowingly subjects another person to any sexual contact without consent commits the offense of sexual assault.” “Sexual contact” is defined in § 452101(66), MCA *14 (2001):
“Sexual contact” means touching of the sexual or other intimate parts of the person of another, directly or through clothing, in order to knowingly or purposely:
(a) cause bodily injury to or humiliate, harass, or degrade another; or
(b) arouse or gratify the sexual response or desire of either party.
¶21 Clark argues that subsections (a) and (b) describe distinct acts and that the District Court should have instructed the jury that it had to be unanimous as to how Clark subjected T.C. to sexual contact. Its failure to do so, Clark contends, was plain error.
¶22 Clark relies on our decisions in
State v. Weaver,
¶23 In
Weldy,
we held that the three subsections of § 455202(2), MCA (1993), defining felony assault, were “not alternative means of satisfying one common element. Rather, they each set forth separate offenses in themselves.”
Weldy,
¶24 Clark’s reliance on
Weaver
and
Weldy
is misplaced. Section 452101(66), MCA, allows for two means of satisfying the “knowingly or purposely” element of sexual contact; it does not articulate two separate offenses.
See Kills On Top v. State
(1995),
¶25 Therefore, we discern no claimed errors that implicate Clark’s *15 fundamental constitutional rights, and we decline to exercise plain error review of the District Court’s unanimity instruction to the jury. ¶26 Did the District Court abuse its discretion by denying Clark’s motion for a new trial after the complaining witness recanted her testimony?
¶27 After consideration of the parties’ arguments, we conclude herein that the state of the law as it relates to motions for a new trial based on new evidence and, more specifically, recantations of trial testimony by a witness, is muddled and in need of clarification.
3
We begin by addressing the test we stated in
State v. Greeno
(1959),
¶28 Several variants of the Berry test are used by courts throughout the nation. The federal courts use different versions to evaluate Rule 33, Fed.R.Crim.P., motions for a new trial based on new evidence. Likewise, many state courts use their own versions of the Berry test for analogous new trial motions. The original test read as follows:
Upon the following points there seems to be a pretty general concurrence of authority, viz.: that it is incumbent on a party who asks for a new trial, on the ground of new discovered evidence, to satisfy the Court, 1st. That the evidence has come to his knowledge since the trial. 2d. That it was not owing to the want of due diligence that it did not come sooner. 3d. That it is so material that it would probably produce a different verdict, if the new trial were granted. 4th. That it is not cumulative only — viz.: speaking to facts, in relation to which there was evidence on the trial. 5th. That the affidavit of the witness himself should be produced, or its absence accounted for. And 6th, a new trial will not be granted, if the only object of the testimony is to impeach the character or credit of a witness.
Berry,
¶29 In 1928, the United States Court of Appeals for the Seventh Circuit in
Larrison v. United States
(7th Cir. 1928),
[A] new trial should be granted when, (a) The court is reasonably well satisfied that the testimony given by a material witness is false, (b) That without it the jury might have reached a different conclusion, (c) That the party seeking the new trial was taken by surprise when the false testimony was given and was unable to meet it or did not know of its falsity until after the trial.
Larrison,
¶30 In
Perry
we outlined yet a different rule to apply in cases where a prosecution witness recants his trial testimony after a verdict is announced and the defendant moves for a new trial on the grounds that the recantation is newly discovered evidence. We stated in
Perry,
When a new trial is sought on the basis of recanting testimony of a prosecution witness, the weight to be given such testimony is for the trial judge passing on the motion for a new trial to determine. The trial judge is required to grant a new trial only when he is satisfied the recantation of the witness is true.
The purpose of the Perry rule was to recognize the district court’s wide latitude in its consideration of a motion for a new trial in this context. However, when considered together with our precedent from Greeno and Greeno’s incorporation of Berry, Perry presents legal and logical problems.
¶31 The legal history of the
Perry
rule reveals a subtle evolution of a sensible rule into an unworkable one. Our rule in
Perry
was taken directly from a case decided by the Supreme Court of Kansas,
State v. Norman
(Kan. 1982),
¶32 This further step illustrates the logical problem of our Perry rule. In the context of a prosecution witness’s recantation that is material to the issues at trial, the Berry element permitting a new trial if the evidence would “probably” produce a different verdict is practically irrelevant because a district court may order a new trial only when it is convinced the recantation is true. If a trial judge *18 believes the recantation to be true, the “probably” requirement becomes woefully inadequate, as the recantation of material testimony would almost certainly produce a different result. Conversely, if a trial judge believes the recantation to be false, how could it “probably” produce a different result? The only situation in which the Perry rule and the Berry rule could effectively work in concert is when a trial judge believes one thing about the veracity of the recantation while simultaneously believing a new jury would probably disagree with him. This small category of circumstances is not one upon which our general principles of law in this area should be focused. Moreover, the plain reading of the Perry rule is troublesome, for it inappropriately places the judge in the role of fact-finder, inevitably, in some cases, on the key matter of the guilt or innocence of the accused. Therefore, to the extent that Perry stands for the proposition that a “trial judge is required to grant a new trial only when he is satisfied the recantation of the witness is true,” it is overruled.
¶33 However, to the extent that Perry stands for the proposition, “When a new trial is sought on the basis of recanting testimony of a prosecution witness, the weight to be given such testimony is for the trial judge passing on the motion for a new trial to determine,” it is retained as authoritative precedent. This consideration, as explained below, should be merely part of the Berry analysis.
¶34 As noted above, the Berry test has seen innumerable alterations over its life, usually with little or no rationale or explanation for the differing versions. We now perceive the Berry test, as articulated in Greeno, to be antiquated and imprecise. Therefore, in an effort to establish a clearer test with a cogent rationale, we restate the Berry test as follows.
To prevail on a motion for a new trial grounded on newly discovered evidence, the defendant must satisfy a five-part test:
(1) The evidence must have been discovered since the defendant’s trial;
(2) the failure to discover the evidence sooner must not be the result of a lack of diligence on the defendant’s part;
(3) the evidence must be material to the issues at trial;
(4) the evidence must be neither cumulative nor merely impeaching; and
(5) the evidence must indicate that a new trial has a reasonable *19 probability of resulting in a different outcome. 6 , 7
¶35 We have dropped the requirement that the new evidence be supported by affidavit because imposing the requirement is at best superfluous and at worst too limiting to a district court’s evaluation of new evidence.
¶36 The fifth element, pertaining to reasonable probability of a different outcome, is most likely to be the crux of any district court’s evaluation of new trial motions based on new evidence. In the present context, “reasonable probability” is somewhere between the Larrison test’s “might have reached a different conclusion” standard and Berry’s “probably produce a different verdict” standard, and for good reason. 8 In any given case, a jury “might” have reached a different conclusion based on any small, even irrelevant, change in trial evidence because “might” means “any chance at all.” This retrospective test is simply too broad. In contrast, a district court could be convinced that the new evidence before it has a strong chance of bringing about a different verdict upon a new trial, but it may not think this possibility so strong that it “probably” would produce a different verdict-i.e., that it has a 51 percent or greater chance of producing an a different verdict. This prospective test is too restrictive. However, the reasonable probability standard adopted herein properly leaves to the trial judge determinations of weight and credibility of the new evidence, and to consider what impact, looking prospectively at a new trial with a new *20 jury, this evidence may have on that new jury.
¶37 In evaluating a motion for new trial under the restated
Berry
test, a district court will naturally consider many factors. In the context of recantations, for example, nothing in our decision today negates the concerns we expressed in
Perry
that recantations are to be “viewed with great suspicion” and that they “demonstrate!!] the unreliability of a witness,” lest the finality of trial verdicts be at the mercy of the scofflaw or unreliable witness.
Perry,
¶38 To summarize, as a district court applies the restated Berry test to a motion for a new trial based on a recantation by a prosecution witness, the district court is not to make factual determinations as to the veracity of the recantation. Rather, the district court should evaluate the recantation for materiality and its tendency to be cumulative or impeaching, and should determine whether there is a reasonable probability that a new trial would produce a different outcome. Keeping in mind the aspects of recantations that make them inherently suspect, determinations of weight and credibility are left to the trial judge. Then, if the Berry elements are properly met, a new trial is warranted. If not, the prior judgment stands.
¶39 A further comment is in order regarding the applicable standards of review. We have held that, generally, we will review rulings on motions for a new trial for abuse of discretion.
See
¶ 18 above. While this remains true, we note that, to the extent that a district court makes findings of fact in its application of the restated
Berry
test (i.e., in relation to elements one and two), those findings must be made by a preponderance of the evidence and will be reviewed for clear error.
*21
Compare
State v. McCallum
(Wis. 1997),
¶40 We turn now to the application of the restated Berry test to the present case. Clark focuses his argument on the Berry elements, contending that the circumstances of T.C.’s recantation satisfy the test. However, he does not address how Perry bears on our evaluation of the District Court’s order denying his motion for new trial.
¶41 The State responds that the new evidence fails the Berry test because T.C.’s recantation is “not so material that it would probably produce a different result” upon a new trial. However, the State relies more heavily on Perry. The State argues that serious problems with T.C.’s credibility were explored at trial, that T.C.’s recantation was motivated by her desire to repair her relationship with her mother, and that T.C. was subject to pressure from Clark, Christina, and others to conform her story to Clark’s. For these reasons, the State argues, the District Court properly exercised its discretion under Perry in denying the motion for a new trial.
¶42 Given our refinement of the Berry test, our partial overruling of Perry, and our inability to glean from the record the District Court’s reasons for denying the motion for a new trial, we cannot discern whether the law has been correctly applied. Therefore, we are obliged to reverse the District Court’s order and remand for the District Court to undertake anew Clark’s motion for new trial.
CONCLUSION
¶43 For the foregoing reasons, we decline to exercise plain error review of the District Court’s unanimity instruction to the jury; thus, we affirm the judgment of conviction. However, we reverse the District Court’s order denying Clark’s motion for a new trial, and we remand the cause for further proceedings consistent with this opinion.
¶44 Affirmed in part, reversed in part, and remanded for further proceedings.
Notes
Only the conviction for sexual assault is challenged on appeal.
Subsection (c) was not at issue.
Section 46-16-702, MCA, permits courts to grant a new trial “if required in the interest of justice.”
See, e.g., State v. Loose
(Utah 2000),
Our statement of the elements was virtually identical to
Berry.
“(1) That the evidence must have come to the knowledge of the applicant since the trial; (2) that it was not through want of diligence that it was not discovered earlier; (3) that it is so material that it would probably produce a different result upon another trial; (4) that it is not cumulative merely-that is, does not speak as to facts in relation to which there was evidence at the trial; (5) that the application must be supported by the affidavit of the witness whose evidence is alleged to have been newly discovered, or its absence accounted for; and (6) that the evidence must not be such as will only tend to impeach the character or credit of a witness.”
Greeno,
For relatively similar restatements of the
Berry
test, see
Kulczyk,
There is inconsistency in the federal courts’
Berry
terminology. Some of the federal courts appear to equate “reasonable probability with “probably,” though the terms have plainly different meanings.
See, e.g., United States v. Rouse
(8th Cir. 2005),
For a useful discussion of the confusion courts have created about the appropriate analysis regarding how new evidence could affect a new jury or, perhaps, how the original evidence may have affected the trial jury, see United
States v. Huddleston
(D. Me. 1998),