(dissenting). I agree with the majority’s rejection of motive, intent, absence of mistake, and bolstering the victim’s credibility as proper MRE 404(b) purposes under the facts of this case. However, I dissent because defendant’s scheme, plan, or system in doing things was not a purpose offered or articulated at trial as required by VanderVliet,1 Crawford,2 and Starr.3 Also, the majority fails to show that the evidence of a common plan in this case is relevant apart from its tendency to show the defendant *74committed the instant crime because he acted in conformity with his criminal disposition. Finally, the majority errs in creating a rule that any similarities among successive crimes proves a plan, and thus proves that the charged offense was committed.
PROPER PURPOSE
VanderVliet teaches that something more than a mechanical recitation of MRE 404(b) list of “proper purposes” is required in order to move forward to the remaining three legs of its test.4 A determination must be made that the purpose claimed is relevant to an issue other than propensity. This protects against the introduction of extrinsic act evidence when that evidence is offered solely to prove character. 444 Mich 74. In this determination, relevance is intertwined with whether a proper purpose was presented. In order to provide a proper purpose under 404(b), the evidence must be relevant to a material fact or issue in the case. That is, the purpose for which the 404(b) evidence is offered must relate to an issue of consequence other than defendant’s arguably MRE 401 relevant propensity to commit certain bad acts. Whether the purpose is proper is a fact-specific, case-by-case determination.
*75It is undisputed that the prosecution did not present the other acts evidence for a proper purpose.5 At trial, the prosecution presented the evidence to show defendant’s motive and intent, his absence of mistake, to demonstrate the possibility that adults can be sexually attracted to children, and to bolster the complainant’s credibility. The trial court made a conclusory determination that the evidence was relevant and admissible because defendant had committed a similar act in the past. In its ruling, the court did not state the purpose on which it relied. It merely stated that the evidence was presented for a proper purpose. When instructing the jury, the trial court warned that the evidence was proper only to show defendant’s scheme, plan, or system in doing certain things.
Each of the prosecution’s arguably plausible6 purposes are properly rejected by the majority. However, the majority has gone one step further, and makes a determination whether “scheme, plan, or system” is a *76proper purpose under these facts. Apparently, this issue was preserved for our review because the trial court instructions to the jury recited these magic words.
Judge Whitbeck’s dissent below properly recognized that the initial question for the appellate court is whether, under VanderVliet, the prosecutor articulated a proper purpose at trial. 236 Mich App 1, 16; 600 NW2d 98 (1999). It is insufficient for the prosecutor, on appeal, to recite purposes that were articulable at trial. Id. Under VanderVliet, the prosecutor has the burden of articulating a nonpropensity ground for admission of other acts evidence. It is not enough that a nonpropensity theory might exist if it was not articulated or relied on by the prosecution at trial. This was reiterated in both Starr and Crawford. See 236 Mich App 17, n 6 (explaining that those cases require that the proper purpose be offered or articulated by the prosecutor, rather than simply requiring that a proper purpose exists). Moreover, it is not enough that the trial court may have listed or articulated a proper purpose when accepting the evidence. The first leg of VanderVliet requires the prosecutor at trial to articulate a proper purpose for admitting the prior acts evidence, it does not look forward to the purposes articulated by the trial court when accepting the evidence.
This view is further supported by this Court’s adoption of a procedural safeguard in these cases after VanderVliet. MRE 404(b)(2) provides that the prosecution “shall” not only provide pretrial notice of its intent to introduce 404(b) evidence, but it must also provide the rationale, whether or not it is mentioned in 404(b)(1), for admitting the evidence. This require*77ment is consistent with our case law because it requires the prosecutor to do more than simply recite the proper purposes listed in MRE 404(b)(1).
Additionally, as a matter of fairness to the defendant, the prosecutor should not be allowed to list its purposes and rationale at trial, and then grasp for more on appeal when those fail. MRE 404(b)(2) provides the defendant an opportunity to respond to the prosecutor’s list of purposes and rationale for admission. If the prosecutor is allowed on appeal to argue that unarticulated purposes existed, we would discourage compliance with this rule.
Judge Whitbeck scoured the record to find any and all arguments showing the prosecution’s nonpropensity theories.7 Scheme, plan, or system was not included in the list. In fact, it was not until the case reached this Court a second time that the “scheme, plan, or system” theory was added by the prosecutor. The prosecutor cites People v Engelman, 434 Mich 204, 223, n 27; 453 NW2d 656 (1990), for the proposition that the failure to advance the correct theory of admission does not justify automatic reversal of a conviction. This fails because, first, this was dicta, and, second, Judge Whitbeck correctly noted that the cases relied upon in the Engelman footnote do not support the prosecutor’s view. Judge Whitbeck stated:
I first note that Engelman, in turn, cites Templin v Not-tawa Twp, 362 Mich 257; 106 NW2d 825 (1961), and Plec v Liquor Control Comm, 322 Mich 691; 34 NW2d 524 (1948) .... Templin, however, does not deal with an evidentiary rule and stands only for the time-honored proposition that a *78reviewing court will not disturb the conclusions of a lower court that reached the right conclusion regardless of the reasons it cites for reaching that conclusion. Templin, supra at 261. The decision in Plec rests on the same basis .... At best, therefore, it seems to me that Engel-man stands for the proposition that the fact that a trial court may have erred in assigning a proper purpose for accepting the other acts testimony is not fatal on appeal, if a proper purpose existed and was articulated at trial. I secondly note that, quite obviously, Engelman predated the Court’s adoption of the VanderVliet four-legged test and therefore should not be interpreted as speaking definitively to the question of an “articulated” versus an “articulable” purpose with respect to the first leg. [236 Mich App 20, n 10 (emphasis in original).]
After two attempts at the Court of Appeals, and two attempts at this Court, it appears that the prosecutor has finally articulated a proper purpose under the majority view. It is apparent that the first leg of the four-legged test articulated in VanderVliet has been rendered a nullity. That case is left to wobble on its remaining three legs. From today forward, prosecutors may ignore the requirements of 404(b)(1) and (2) along with the first requirement of VanderVliet. On appeal, they may list any and all proper purposes in an attempt to hit the mark with one, or they may allow the appellate court to find one for them, thereby affirming a criminal conviction.
Because the prosecutor failed to meet the requirements of the first leg of VanderVliet and our court rules regarding character evidence, I would affirm the decision of the Court of Appeals.
SCHEME, PLAN, OR SYSTEM
As explained, the proper purpose articulated by the prosecutor must be relevant to a fact in issue. The *79majority properly analyzes this requirement in its discussion of the prior bad acts to show defendant’s motive, intent, absence of mistake, and bolstering the victim’s credibility. However, its analysis of scheme, plan, or system is lacking this requirement.
The majority explains that the proper purpose must be relevant, or material:
Materiality, however, “does not mean that the evidence must be directed at an element of a crime or an applicable defense.” A material fact is one that is “ ‘in issue’ in the sense that it is within the range of litigated matters in controversy.” [Ante, p 57 (citations omitted).]
In finding that defendant’s scheme, plan, or system was material, the majority relies on an interesting interpretation of the Engelman dicta. Ante, pp 61-62. It is true that other acts evidence is not limited to proving identity or intent. We need not look to Engel-man to support this statement when 404(b) itself provides a nonexhaustive list of the proper purposes for admitting evidence. However, the propositions surrounding the Engelman quote are simply inaccurate. The majority provides that “We [this Court] held that evidence of other instances of sexual misconduct that establish a scheme, plan, or system may be material in the sense that the evidence proves that the charged act was committed.” Ante, pp 61-62. To the contrary, in Engelman we held that photographic evidence of the defendant standing nude next to a minor female was inadmissible to show that defendant had a scheme, plan, or system of luring children into his home for sex acts even though defendant had taken nude photos of the victim in that case. 434 Mich 221. The holding was based on a lack of a “true plan.” Id. The instant majority provides that Engelman “sug*80gested” that a trae plan required that defendant formed a plan including the charged and uncharged crimes as stages in the plan’s execution. Ante, pp 62-63. In fact, Engelman held that because a true plan was not shown, as defined by Imwinkekied, Uncharged Misconduct Evidence, § 3:21, p 53, and 1 Wharton, Criminal Evidence (14th ed), § 186, pp 786-787, the evidence was inadmissible. 434 Mich 221. Finding no support in our case law,8 the majority looks to other jurisdictions to add a new theory of logical relevance that will fit within the category of scheme, plan, or system. It relies on State v Lough, 125 Wash 2d 847; 889 P2d 487 (1995), and People v Ewoldt, 7 Cal 4th 380; 867 P2d 757 (1994). However, the theory derived from those cases does not “clarify” our case law, but instead rejects it.
Lough simply followed the Ewoldt analysis and its precedent. The Ewoldt rule provides that other acts *81evidence is relevant to prove that the charged act was committed if it is sufficiently similar to support the inference of a common design or plan. 7 Cal 4th 402. This view was rejected in Engelman, where this Court explained that when the ultimate fact to be proved through the other acts evidence is that the criminal act took place, the question becomes whether the evidence offered tends to establish some intermediate inference, other than the improper inference, which is probative of the ultimate issue in the case — the commission of the act. 434 Mich 217. In other words, there must be a proper intermediate inference established by the other acts evidence that is probative of whether the crime occurred. Id. at 218.
The majority has provided no intermediate inference.9 Instead, it adopts the rule that the prosecutor need only show similarities in the charged and uncharged crimes. These similarities prove only that defendant acted in conformity with his character to commit similar acts. This analysis allows the prosecution to skip the intermediate inference requirement and instead point to one or two similarities as proof that the second act must have occurred. In other words, the prosecutor may now work backwards and argue that, where there are similarities, there is a plan, and therefore the other acts evidence proves defendant committed the instant crime. This logic has never been adopted, and it appears this leap is now made simply to convict a reprehensible defendant. The proper rule would follow well-established principles of evidence that require a showing that the *82other acts evidence is being presented to show something other than propensity to commit the crime. Crawford, supra at 390.
SUFFICIENT SIMILARITY
The majority’s new rule removes the prosecutor’s burden of weaving a logical thread linking the prior act to the ultimate inference. Rather than follow our established law in this area, it adopts California’s “sufficient similarity” test provided in Ewoldt. However, under the facts of this case, it fails to carry even this light burden. Ewoldt explains that to establish a common plan, the common features must indicate the existence of a plan rather than a series of similar spontaneous acts, but the plan need not be distinctive. Ewoldt, supra at 403.
In this case, the majority points to three common features among the uncharged and charged acts. First, the alleged victims had a father-daughter relationship. Second, the victims were of similar age at the time of the abuse. Third, defendant played on his daughters’ fears of breaking up the family to silence them. Ante, p 66. The problem with such common features is that they include the very elements of the uncharged and charged crimes. Defendant was convicted of CSC I pursuant to MCL 750.520b(l); MSA 28.788(2)(1). That statute provides that defendant is guilty if he engages in sexual penetration with another person, that other person is at least thirteen but less than sixteen years of age, and any of the following: (i) the actor is a member of the same household as the victim; (ii) the actor is related to the victim by blood or affinity to the fourth degree; (iii) the actor is in a position of authority over the victim and *83used this authority to coerce the victim to submit. In this case, both victims were between the ages of thirteen and sixteen at the times of the offenses, the majority’s second similarity. Defendant was a member of the same household as both victims, and defendant was related by either blood or affinity to both victims, the majority’s first similarity of a father-daughter relationship. Finally, as a father or stepfather he was in a position of authority over both victims and arguably used it to get them to submit, which is the majority’s third similarity of use of threats.10
It is obvious that the preceding logic will do great damage to the law of evidence if the prosecutor is allowed to introduce the very elements or nature of a prior crime in order to show that the defendant committed a later crime. Any second offender of a charged or uncharged similar crime will have his prior acts disclosed to the jury in order to prove a plan. There is no longer a requirement that the prosecutor show a permissible intermediate inference to prove any material fact. Similarities instantly show a plan, and use of a plan shows that defendant committed the crime at issue.
Moreover, the majority has shown nothing to indicate that the common features of the charged and uncharged crimes “indicate [d] the existence of a plan rather than a series of similar spontaneous acts” under the Ewoldt test. Id. at 403. Rather, the facts indicate that defendant spontaneously took advantage of an opportunity to rape his daughter while they *84were alone in the house. No facts show that he manipulated the situation in order to get his daughter alone in the house as part of a plan to later attack her. The threat that she would break up the family appears to have been an afterthought to cover up the crime after it occurred rather than part of an overall common plan to commit CSC I. Finally, no facts link the prior acts against defendant’s stepdaughter to a plan to rape defendant’s daughter ten years later. The majority has failed to even show the impermissible “series” of spontaneous acts. While it is alleged that defendant committed a series of similar acts against his stepdaughter, the instant offense is remote in time and cannot be considered a part of that series. Instead, the instant offense is nothing more than a similar act in that it too is a CSC I offense.
CONCLUSION
The majority errs for several reasons. First, it has removed the responsibility of the prosecutor to provide and articulate a proper purpose as required under the first leg of the VanderVliet test. Instead, it adopts a new “common plan” purpose not argued by the prosecutor. This allows appellate courts, in hindsight, to search for proper purposes to support admission of MRE 404(b) evidence despite the fact that defendant was not allowed to argue against its admission. Because the prosecutor failed to carry its burden for introduction of the other acts evidence, our inquiry should end. Second, the majority errs in removing the longstanding requirement of showing that the evidence of the plan legitimately and logically shows that defendant committed the crime. The majority simply asserts that the commission of the *85crime is the ultimate fact, but fails to explain how the facts of the first sex offense led to the conclusion that the second offense was committed. The majority cannot show that the evidence of a common plan in this case is relevant apart from its tendency to show that the defendant committed the instant crime because he acted in conformity with his criminal disposition. Third, the majority errs in creating a rule that any similarities among successive crimes proves a plan, and thus proves that the charged offense was committed. Even under that test, the similarities ought not be the elements or nature of the crime itself.
For the above reasons, I would affirm the decision of the Court of Appeals.
Kelly, J., concurred with Cavanagh, J.People v VanderVliet, 444 Mich 52; 508 NW2d 114 (1993).
People v Crawford, 458 Mich 376; 582 NW2d 785 (1998).
People v Starr, 457 Mich 490; 577 NW2d 673 (1998).
The VanderVliet test provides that, first, evidence must be relevant to an issue other than propensity under Rule 404(b); second, the evidence must be relevant under Rule 402 as enforced through Rule 104(b) to an issue of fact or consequence at trial; third, the trial court should employ a balancing test under Rule 403 to determine whether the probative value of the similar acts evidence is substantially outweighed by its potential for unfair prejudice; and, fourth, the trial court, upon request, may provide a limiting instruction under Rule 105.
It is not solely the prosecutor’s failure to recite magic words that requires reversal in this case. It is the failure under our case law and MRE 404(b)(1) and (2) to provide a legitimate rationale for admitting the evidence. The majority argues that pointing out similarities between an uncharged and charged act satisfies the prosecutor’s burden to articulate and provide rationale for admission of the evidence. In future cases, criminal defendants should cite this rationale to ensure review of issues not preserved at trial. The majority also argues that the trial court admitted the evidence under a “scheme, plan, or system” theory. The record is void of support for this assertion. The court simply stated that the evidence was relevant because it was similar. Finally, instructing the jury that it should use the evidence to support a showing that this is “how he [defendant] does certain things” fails to cure the error.
As the amicus curiae Criminal Defense Attorneys of Michigan brief points out, the suggestion that other acts evidence is relevant to show that adults can be sexually attracted to children is nothing more than an ingenious disguise of an improper propensity purpose. No rational juror in our society today would doubt that an adult could be sexually attracted to a child. The fact that the majority chose not to address the question evidences its agreement that this is implausible.
Judge Whitbeck also noted that the prosecution added “modus oper-andi,” “corpus delicti,” and “corroboration” as new theories at the Court of Appeals. 236 Mich App 19-20, n 10.
The majority’s injection of People v Oliphant, 399 Mich 472; 250 NW2d 443 (1976), fails to provide support for its view. First, Oliphant was decided before we adopted the requirements of MRE 404(b)(2), Vander-Vliet, Starr, and Crawford. Also, in Oliphant, the evidence was material to the issue of consent. In the instant case, the evidence was admitted to prove that the instant offense occurred. As Engelman explains:
“When the prosecutor offers uncharged misconduct to prove the commission of the actus reus, the judge should very carefully scrutinize the prosecutor’s theory of logical relevance. This is the theory of relevance in which it is easiest for the prosecutor to slip into improper character reasoning. Since the ultimate inference is conduct, this theory places the greatest strain on the prohibition in the first sentence of Rule 404(b). The courts are reluctant to admit uncharged misconduct to prove the actus reus, and that reluctance is well-founded. If the prosecutor wants to avoid Rule 404(b), the prosecutor must persuade the judge that the prosecutor is invoking an intermediate inference other than the defendant’s subjective character.” [434 Mich 215-216 (citations omitted).]
Of course, after today, the prosecutor need not bother with these unduly strict requirements, and may introduce evidence of prior acts by pointing to similarities between the prior act and the offense charged.
The majority’s explanation, that the “common system” theory itself is its own intermediate inference, is vacuous, but would fit quite well into a work of Lewis Carroll.
The majority concedes that the charged and uncharged acts were dissimilar in many respects. Rather than acknowledge that the two crimes are separate spontaneous acts lacking a common thread other than the statutory elements, it takes cover in the abuse of discretion standard of review — a standard that now appears insurmountable.