(dissenting) I dissent because the majority has flouted the express language of a legislative enactment and has subverted the legislature’s will. Because the majority apparently feels uncomfortable with the result that following the legislative direction would compel, it capriciously sets that direction aside and embarks upon a discursion into a realm of fantasy which produces a conclusion that satisfies the majority’s preference in this particular case.
The legislature has explicitly stated that, in sexual assault cases, “any evidence concerning the complaining witness’s prior sexual conduct or opinions of the witness’s prior sexual conduct and reputation as to prior sexual conduct shall not be admitted into evidence . . .” subject to three enumerated exceptions. Sec. 972.11(2) (b), Stats. The purposes of such rape shield laws include protecting the complaining witness from public exposure of her private life and preventing the trial from focusing on the prior sexual experience or lack of experience of the complaining witness rather than upon the facts of the incident in question.
The majority states that “[ejvidence relating to a complainant’s prior sexual conduct” may be admitted if it serves to “prove a fact independent of the complain*167ant’s prior sexual conduct which is relevant to an issue in the case . . and two further conditions are satisfied. (Supra at 157) According to the majority, evidence of prior sexual conduct may be admitted if the trial court determines that it is relevant to an “independent” issue in the case and that it is not excessively prejudicial. This is a judicial fabrication that flies in the teeth of the legislative enactment. The majority’s pronouncement is contrary to the language of sec. 972.11 (2) (b), Stats., which provides that evidence concerning the complainant’s prior sexual conduct shall not be admitted. The acceptance of this unfounded standard, created out of whole cloth by the majority, will defeat the very purposes of the rape shield law.
The majority asserts that the criteria for admission which are intended to apply only to the legislatively ordained exceptions to the rape shield law are to be applied indiscriminately to all proffered evidence of prior sexual conduct. It is impossible for this writer to conjure up any justification for such an erroneous reading of the statute.
A perusal of the statute reveals that it is only the legislatively ordained exceptions to the rule of absolute inadmissibility, which appear in sec. 972.11(2) (b), Stats., that are subject to the rule that the trial judge shall determine admissibility on the basis established by sec. 971.31(11) — that the proffered evidence is in respect to a material fact and “of sufficient probative value to outweigh its inflammatory and prejudicial nature . . .”
Thus, it is apparent that a trial court has, if it abides by the test of sec. 971.31(11), Stats., the discretion only to admit evidence of (1) the complaining witness’ past conduct with the defendant; (2) specific instances of sexual conduct showing the source or origin of semen, pregnancy, or disease, for use in determining the de*168gree of sexual assault or the extent of injury suffered; and (3) prior untruthful allegations of sexual assault made by the complaining witness.
These are the only exceptions that the legislature has permitted to the rule that any evidence of prior sexual conduct is inadmissible. And even these exceptions are only admissible when the trial court has in its discretion concluded that the materiality and probativeness of the evidence outweighs the prejudice that would follow the admission of the evidence. Yet by a process not explained, the majority jumps to the unwarranted conclusion that any evidence of prior sexual conduct may be admissible if, in the discretion of the trial court, the probativeness of the evidence outweighs its prejudicial effect. This is incorrect, because the majority ignores the fact that the legislature has stripped a trial judge of all such discretion except that which may be exercised in respect to the types of prior sexual conduct that are made conditionally admissible by sec. 972.11 (2) (b). Discretion may be exercised only in respect to those specified types of evidence. Even the exceptions are to be rigorously tested before they are deemed to be admissible. All other evidence of prior sexual conduct is barred absolutely.
The statute represents a considered legislative determination that prior sexual conduct evidence, subject to limited stated exceptions, is irrelevant or, if relevant, that its probative value, ipso facto, is outweighed by its prejudicial effect. In Milenkovic v. State, 86 Wis. 2d 272, 279, 272 N.W.2d 320 (Ct. App. 1978), the court of appeals acknowledged that courts had, at an earlier time, concluded that a woman who is “previously unchaste” is more likely to consent to an act of sexual intercourse than is a woman who is “strictly virtuous.” The court rejected that view. It also rejected the view that prior sexual conduct is relevant to credibility.
*169The majority notes that sec. 972.11(2) (b), Stats., is “a legislative determination that ordinarily evidence of a complainant’s prior sexual conduct has low probative value but high prejudicial effect.” (Supra at 157, n. 2) The statute actually shows that the legislature determined that evidence of prior sexual conduct has such low probative value and such a highly prejudicial effect that it must be excluded except in regard to the enumerated circumstances. The majority does not attack this legislative determination, yet it allows such evidence to be admitted even when those circumstances are not present.
Evidence of prior sexual conduct was never, and could never be, admitted simply to prove prior sexual conduct, because that is never an issue in a sexual assault case. It was admitted only because courts at one time thought it was relevant to issues such as consent and credibility. The rape shield law is a legislative declaration that such evidence is not admissible, no matter to what issue the trial court believes it is relevant.
Under the majority’s formulation, evidence of prior sexual conduct can once again be admitted by trial courts at the request of the state or, as the majority fails to perceive, at the request of the defendant. This substantially nullifies the protection which the legislature intended to afford rape victim complainants. It will also result in the admission of evidence which tends “to afford the trier of the fact an opportunity for distraction that suggests abandoning the duty to consider the evidence.” Milenkovic, 86 Wis. 2d at 281.
In this case, none of the exceptions permitted by sec. 972.11(2) (b), Stats., were present. Therefore, evidence of the complaining witness’ prior sexual conduct was inadmissible. The trial court, as the result of the legislature’s mandate, had no discretion to exercise. The evidence was excluded by statute.
*170The majority properly finds error in the admission of evidence by the complaining' witness that she was a virgin before the incident and in the testimony by Dr. Hilrich that, in his opinion, the complainant was a virgin prior to this incident. However, it finds that the doctor’s testimony concerning his observation of a recent tear in the complainant’s hymen would have been admissible if a proper limiting instruction had been given.
The majority acknowledges that the doctor’s testimony concerning the hymen tear was evidence of prior sexual conduct. That is true in this case because Dr. Hilrich made it clear that he believed the tear resulted from the complainant’s first experience of sexual intercourse. The majority states, however, that the testimony relating to the hymen tear was evidence of force and, thus, was relevant to the issue of consent.
The hymen may be torn by consensual or non-consensual penile penetration, as well as by other causes. There is no evidence of record that this particular hymen tear was more likely caused by non-consensual than consenual penetration. Therefore, evidence of the hymenal tear was not relevant to the issue of consent. Dr. Hilrich testified that, from his observations, he could say that there was forceful entry into the vagina. He did not specify that this conclusion was based on the nature of the hymen tear rather than on the hematoma and the swelling of the labia which he also observed.
On cross-examination, Dr. Hilrich testified that the only way a hymen can be torn is by forceful dilatation, that is, inserting something larger than the opening. He also indicated that first intercourse usually ruptures the hymen. Unless first intercourse is usually non-consensual, this testimony means that first intercourse is usually “forceful” because it involves the insertion of an object larger than the opening in the hymen, which usually results in a tear. Apparently, Dr. Hilrich would *171expect any first intercourse to result in evidence of what he called “forceful entry.” Thus, his testimony did not mean that the defendant forced the complainant to have intercourse. Even if evidence of a hymen tear, arguendo, might be relevant in another sexual assault case, in this case it was not shown to be relevant.
The complainant’s accounts of her conversations with the defendant stand on a different ground. They are properly viewed not as evidence of prior sexual conduct, but as evidence of what took place between the individuals in respect to showing consent or lack of consent. A person faced with unwanted sexual advances may do or say a variety of things to avoid the attack. A woman might tell her assailant, for example, that she was pregnant, or had venereal disease, or that she was a virgin —whether or not these claims were true — in order to discourage a perceived threat. This type of statement may be admissible. The trial judge must, of course, under the rules that apply to the admissibility of all potentially prejudicial evidence, determine that it is relevant to consent and that its probative value outweighs its prejudicial effect. Because a jury may draw conclusions about the complainant’s prior sexual conduct from such testimony, a cautionary instruction must be available to restrict the testimony to its proper scope. Such an instruction was not given in this case.
The trial court erroneously admitted evidence of the complainant’s prior sexual conduct consisting of testimony by the complainant and a doctor that she was a virgin and testimony by the doctor about the hymen tear. This evidence was highly prejudicial. The court also erred by failing to give proper limiting instructions in respect to the conversation between the complainant and the defendant.
The prejudicial nature of Dr. Hilrich’s testimony was emphasized by the trial court’s “limiting instruction.” *172The court instructed the jury that the testimony concerning virginity was admitted to show the complainant’s physical condition as it may have bearing on her state of mind. This encouraged the jury to make the impermissible inference that the complainant’s virginity — her physical condition — was relevant to whether she consented — her state of mind.
The prejudicial effect of the virginity evidence was magnified by the remarks of the prosecutor in his opening and closing statements.1 In the opening statement, he said that the complainant was a virgin and that the person that had intercourse with her “tore her open.” In closing argument, the prosecutor said the complainant was naive and he referred to her virginity several times. He questioned whether she would have consented in the situation the defendant described saying, “[a]nd on the basis of that kind of persuasion, she sacrifices her virginity that she for some reason had maintained for 23 years. Your virginity may not have meant much to you, but I think it means quite a lot to her, and it did.” Again, on rebuttal, the assistant district attorney asked the jury to consider the reasonableness of the defendant’s argument, “This woman — she’s 23 years old, a virgin. She’s sleeping in her apartment. She hears knock, knock, knock down the hall and gets up and decides, ‘It's time to lose my virginity.’ ”
This argument actively encouraged the jury to draw the forbidden inference that the complainant’s virginity meant that she did not consent. It brought the virginity evidence to the jury’s attention, increasing the probability that the jury used the virginity testimony as evidence of the complainant’s credibility. The majority simply ignores the inflammatory statements of the prose*173cutor that were rather clearly calculated to prejudice the jury on the basis of inadmissible evidence.
As the majority points out, the credibility of the defendant and the complaining witness was critical to this case. The jury had to choose between two stories. The complainant admitted that she allowed a strange man into her apartment at about 4:00 a.m., that it was her idea that the defendant sleep on her couch, and that she allowed her clothing to be removed without objection. This is not to say that the complainant necessarily consented to sexual intercourse, but the complainant’s own testimony could raise substantial doubts in the mind of a reasonable and uncontaminated jury.
The majority concludes that the physical evidence corroborates the complainant’s testimony to such an extent that the errors in this case were harmless. The physical evidence, however, is far from conclusive. As mentioned above, when Dr. Hilrich testified that the physical evidence showed there had been forceful entry into the vagina, he did not state that, therefore, the entry was non-consensual or different from a typical first intercourse.
Dr. Hilrich testified that red marks on the complainant’s neck were bruises, which was consistent with the complainant’s story. On the other hand, a police officer, who interviewed the complainant shortly after the incident, testified that the marks looked like “hickeys” to her and that the complainant said she thought they came from the defendant kissing her on the neck.
The key issue in this case was consent. The physical evidence was not strong. Some of the complainant’s admitted actions could raise doubts about her story. This was a close case. In a short trial, the jury heard highly prejudicial testimony from two sources. The prejudicial impact of this testimony was increased by the judge’s erroneous instruction and by the improper argument of *174the prosecutor. The jury very likely decided the case on extraneous factors not relating to the evidence in respect to the assault itself and in the legally mistaken belief that the complainant’s credibility was enhanced by the fact that she was a virgin, and the feeling inculcated by the prosecutor that, because she was a virgin, she did not consent.
The defendant was prejudiced. The Wold test is not whether there was sufficient evidence to convict without the tainted evidence, but whether the jury would have done so.2 The court of appeals was correct in concluding that the result at Gavigan’s trial might probably have been favorable to him had the virginity evidence been excluded. Again, it should be emphasized that the majority’s holding in this case opens the evidentiary door for the revelation of the prior sexual conduct of a person who complains of a sexual assault. The protection that the legislature sought to afford a rape victim is substantially diminished by the majority’s opinion.
I would affirm the court of appeals and reverse the trial court and remand for a new trial.
1 am authorized to state that CHIEF JUSTICE BEIL-FUSS and JUSTICE ABRAHAMSON join in this dissent.
Section 972.11(2) (b), Stats., also forbids references to the complainant’s prior sexual conduct.
Wold v. State, 57 Wis. 2d 344, 356, 358, 204 N.W.2d 482 (1973). While the majority correctly quotes Wold at 356, it misapplies the Wold test. This court in Wold clearly applied the test of harmless error in respect to its effect upon a jury and did not confine its scrutiny solely to the effect that the untainted evidence would have as regarded solely from the viewpoint of an appellate court.