State v. McMahon

519 N.W.2d 621 | Wis. Ct. App. | 1994

186 Wis.2d 68 (1994) 519 N.W.2d 621

STATE of Wisconsin, Plaintiff-Respondent,
v.
John E. MCMAHON, Defendant-Appellant.^[†]

No. 93-2412-CR.

Court of Appeals of Wisconsin.

Submitted on briefs April 26, 1994. Decided June 15, 1994.

On behalf of the defendant-appellant, the cause was submitted on the briefs of Norman L. Goeschko of Love, Voss & Murray of Waukesha.

On behalf of the plaintiff-respondent, the cause was submitted on the brief of James E. Doyle, attorney general, and G.M. Posner-Weber, assistant attorney general.

Before Anderson, P.J., Brown and Snyder, JJ.

BROWN, J.

John E. McMahon was charged with nine sexually related offenses involving his first cousin, Nannette S., a minor, and was convicted of six of them. He raises eight issues on appeal. We reverse one count charging sexual intercourse because there was no evidence of penetration. In all other respects, we either find no error or harmless error.

The facts will largely be set forth as they pertain to the individual issues. It is enough to know at the outset that the nine counts spanned a time period from May 1986 through July 1991. The charges grew out of a discussion Nannette had with a social worker from the Waukesha County Human Services while Nannette was in a foster home. After Nannette confided to the social worker that McMahon had sexually abused her, the two of them went to the Waukesha police where Nannette gave a statement. Based upon this information, the State charged McMahon with the nine counts. These charges covered a five-year span beginning when Nannette was in sixth grade and continuing into the eleventh grade. They involved allegations of oral sex, sexual intercourse and vaginal insertion.

Prior to trial, Nannette recanted her statement and claimed that the sole reason for accusing McMahon was because she knew that her mother, McMahon's aunt, did not like McMahon and she felt unloved and discarded by her mother. She claimed that the accusations were made in order to try to please her mother and be closer to her. At trial before the jury, she confirmed her recantation. The State put into evidence her prior inconsistent statements through the testimony of a detective and produced an expert witness on recantation by sexually abused children. The detective also testified that he talked to McMahon and obtained a general admission that oral sex and sexual intercourse did occur at some time. The jury acquitted McMahon of three charges and convicted him of the rest. We now turn to the specific issues.

SUFFICIENCY OF EVIDENCE REGARDING SEXUAL INTERCOURSE IN COUNT NINE

[1]

Count nine charged McMahon with committing an act of incestuous sexual intercourse against Nannette pursuant to § 948.06(1), STATS. This act was alleged to have occurred between July 4 and July 11, 1991. In order to satisfy all of the elements of the offense, the State had to prove that McMahon had sexual intercourse with a child he knows is related and that the child is in fact related by a degree of kinship closer than a second cousin.

Sexual intercourse is defined as:

vulvar penetration as well as cunnilingus, fellatio or anal intercourse between persons or any other intrusion, however slight, of any part of a person's body or of any object into the genital or anal opening either by the defendant or upon the defendant's instruction. The emission of semen is not required.

Section 948.01(6), STATS. The vulva consists of a female's external genitalia. See State v. Morse, 126 Wis. 2d 1, 5, 374 N.W.2d 388, 390 (Ct. App. 1985). Thus, proof of vulvar penetration—however slight—was an essential element of this charge against McMahon. McMahon claims that the trial evidence was insufficient for the jury to find beyond a reasonable doubt that he had sexual intercourse with Nannette during the time alleged in count nine.

Following our independent review of the record, we conclude that McMahon is correct. In her original statement, Nannette told the detective that she had come back to Waukesha from a camping trip and was at another person's home. She called McMahon to see if he could give her some cigarettes. McMahon replied that she should come over right away. Upon arriving, McMahon hugged her, took her by the arms to the couch and pulled her pants down and his halfway down. He got on top of her and tried to have sexual intercourse with her. However, he could not get an erection so he ended up rubbing his penis against her vagina. He told her that he was worried because a friend might be arriving. There is nothing in Nannette's original statement suggesting even slight penetration so as to support the charge.

When Nannette testified at trial, she recanted her prior statement. When asked by the prosecutor if intercourse occurred regarding the circumstances detailed in count nine, she answered, "I don't know." That is the extent of the evidence related to this count.

The State cites Nannette's original statement and McMahon's unspecific admission of sexual intercourse with Nannette to posit that a reasonable jury could find vulvar penetration. In the State's view, the jury could apply its individual and collective experience to reasonably conclude that McMahon's "rubbing" of his penis against Nannette's vagina included slight penetration. The State contends that to conclude otherwise insults common sense.

[2]

We disagree. It is speculative to assume that a flaccid penis rubbed against a vagina will cause penetration—however slight. There is nothing in practical human experience which would permit a juror to conclude that slight penetration occurred based on the meager record. While in the course of human experience an erection is not a necessary prerequisite to slight penetration, the converse is not true—that slight penetration will occur regardless of erection. The truth of the matter, whether penetration occurred, depends upon the facts in each case where penetration is alleged despite the lack of an erection. Here, the facts are simply lacking. We conclude that the evidence was insufficient and reverse count nine.

RIGHT TO A UNANIMOUS JURY VERDICT IN COUNT EIGHT

In Nannette's original statement, she discussed a period of time where she temporarily worked after school making telephone solicitations for vacuum cleaner sales. Nannette originally claimed that, during this approximately one-and-one-half month period, McMahon picked her up about fifteen times while she was walking home from work. She stated that while McMahon took her directly home three of these times, the rest of the time he parked with her before taking her home and had sex with her. On six of those occasions, he parked at a specific location. On each of those six occasions, McMahon inserted his finger in her vagina and she performed oral sex on him. She further indicated that, "some of these times," he "may have given me oral sex and had intercourse with me in the car." She indicated that she could not separate one time from the other in her mind.

The State charged McMahon with one count of incestuous sexual intercourse occurring between October 10 and November 20, 1990. In fact, the exact wording of count eight was as follows:

Between October 10, 1990 and November 20, 1990, the defendant did have sexual intercourse with a child he knows is related by blood in the degree of kinship closer than second cousin, to wit—: female, White juvenile N.S. (D.O.B. 8-4-74) contrary to Section 948.06 (1), Wisconsin Statutes.

The count's wording, on its face, gave no notice that more than one act of sexual intercourse was being charged relative to the time period mentioned.

McMahon argues that, at trial, the State lumped together in a single count the "sheer number of incidents" which allegedly took place during the time period that Nannette held an after-school job. McMahon complains that this amounted to an improper joining of two or more separate criminal offenses. McMahon cites State v. Lomagro, 113 Wis. 2d 582, 586, 335 N.W.2d 583, 586-87 (1983), for the proposition that duplicitous charging may deprive a defendant of jury unanimity. McMahon appears to assert that, since there were twelve incidents spread out over one and a half months, the jurors could disagree as to whether one incident occurred as opposed to another, but still agree that "some" incident occurred. McMahon argues that allowing the jury to do this is constitutionally improper.

[3]

McMahon's trial counsel never objected to the count on duplicity grounds. Therefore, the issue is waived. However, McMahon argues that his trial counsel was ineffective for not having made the motion. Trial counsel admitted during the postconviction hearing that he never considered a motion based on duplicity. We address the duplicity issue within the context of McMahon's ineffective assistance of counsel argument.

[4, 5]

We agree with the parties' assessment of the standards used in determining ineffective assistance of counsel. The standards are as follows: Before any of his convictions can be overturned, McMahon must show that trial counsel's performance was deficient and that his deficient performance prejudiced the defense. Strickland v. Washington, 466 U.S. 668, 687 (1984). The test for deficient performance is whether counsel's representation fell below objective standards of reasonableness. Id. at 687-88. McMahon must identify the specific acts or omissions that form the basis of ineffective assistance. Id. at 690. This court then determines whether, under the circumstances, the identified acts or omissions were outside the wide range of professionally competent assistance. Id. This court presumes that counsel's performance was satisfactory. Id. at 689. We do not look to what would have been ideal, but rather to what amounts to reasonably effective representation.

Regarding the prejudice prong, we will reverse if our confidence in the outcome is doubted such that the conviction is fundamentally unfair or unreliable. Id. at 687.

We use Lomagro as our guide in answering the duplicity issue. The Lomagro court asked: "If the prosecutor issues only one charge but introduces evidence of multiple acts which separately constitute the criminal offense charged, must the jurors unanimously agree as to which act or acts the defendant committed in order to find the defendant guilty?" Lomagro, 113 Wis.2d at 590, 335 N.W.2d at 588. The Lomagro court implied that the question could not be answered with a simple "yes" or "no," but depended upon a case-by-case analysis as follows.

[6]

The first step is to determine whether the jury has been presented with evidence of multiple crimes or evidence of alternate means of committing the actus reus element of the crime. Id. at 592, 335 N.W.2d at 589. If multiple counts are presented to the jury, unanimity is required as to each. Id. If there is only one count presented to the jury, unanimity is not required so long as the multiple acts within the charged offense are "conceptually similar." Id.

Based upon Lomagro, all the acts complained of by Nannette in her original statement were conceptually similar. This is because vulvar penetration, fellatio, cunnilingus and sexual intercourse are all forms of sexual intercourse. The Lomagro court acknowledged that a reasonable person could believe that penis-vagina intercourse and fellatio are conceptually distinct, since the acts involve different orifices of the victim. Id. at 593, 335 N.W.2d at 590. The court, however, held that the legislature performed the conceptual grouping as a matter of public policy. Id. Thus, under Lomagro, all the acts occurring during the time frame of count eight must be considered by us to be conceptually similar. As such, the fact that McMahon was alleged to have engaged in different types of sexual intercourse does not make count eight one that involves multiple acts. The jury need not agree on which act of sexual intercourse occurred; only that an act of sexual intercourse occurred.

This, however, does not end the analysis. While the Lomagro court ruled that the difference in the type of sexual "intercourse" occurring is of no moment, it went on to discuss whether the number of separate acts is a significant consideration.

The Lomagro court held that so long as the multiple acts are all part of "one continuous criminal transaction," it does not matter that multiple acts occurred. See id. at 589, 335 N.W.2d at 588. Thus, for example, the defendant in Lomagro performed three sexual acts within the confines of a car over a two-hour period. Id. at 595, 335 N.W.2d at 591. The Lomagro court compared the facts in the case before it with the facts in Boldt v. State, 72 Wis. 7, 38 N.W. 177 (1888). In Boldt, the State introduced evidence that the defendant, who was charged with selling liquor without a license, sold to different people on different days. Id. at 13, 38 N.W. at 178-79. The Lomagro court distinguished Boldt by writing that Boldt did not involve a single, continuous criminal transaction. Lomagro, 113 Wis. 2d at 596, 335 N.W.2d at 591.

Thus, to determine whether there was a single, continuous criminal transaction, the time between criminal acts is deemed crucial, as is whether the acts involved the same parties. Here, the acts occurred over a month-and-one-half time span. We can find no Wisconsin case, and the State has not cited any, where multiple sexual acts occurring over a time span like the one occurring in the instant case amounted to a single, continuous transaction. We note, however, that the other prong—whether the acts involved the same parties—benefits the State, although the State makes no argument about this.

The State asserts that another consideration in the "one continuous transaction" analysis is whether the time period sets "temporal boundaries." It points to the "temporal boundaries" in count eight as evidence of one continuous transaction set within those boundaries. We find nothing in Lomagro or other cases to support this as a criterion in and of itself, and we think we know why. The fact that a time period sets "temporal boundaries" does not solve anything. The question still remains whether a jury could be in dispute about which of the twelve episodes occurred within the temporal boundaries.

The State then notes that the Lomagro court found as, critical the fact that the defendant flatly denied engaging in any sexual acts with the victim, while the victim maintained that he perpetrated all the acts. In the Lomagro court's view, the question before the jury was simply whether to believe the defendant or the victim. If the jury believed the victim, then the jury believed all of the acts the victim testified to. If the jury believed the defendant, then the jury would disbelieve the victim in whole.

The State posits that the same situation exists here. McMahon has denied all the incidents. Nannette's original statement claimed that twelve such incidents occurred. The State appears to argue that, similar to Lomagro, the jury was asked to believe Nannette's original statement or disbelieve it. If the jury believed Nannette's original statement, they believed all of it as one continuous weaving transaction or story—a connected web surrounding the time period that Nannette briefly worked after school. If they disbelieved it, they believed that no weave and fabric existed.

We are not certain that the State's reading of Lomagro is correct. Our concern is that this is not the same fact situation as existed in Lomagro. In Lomagro, the continuous events occurred over a time span of three hours. Lomagro, 113 Wis. 2d at 583-84, 335 N.W.2d at 585-86. The Lomagro jury was faced with an "either-or" situation within that short time span. During that short time span, the victim recounted what can be described as one continuous, horrible nightmare. Here, the events occurred over a month and one-half. We are unconvinced that Lomagro stands for the proposition that so long as a victim tells a story of assault by one perpetrator, over a set period of time, the law can consider it as one continuous story with various chapters. We think it just as likely that the Lomagro court meant to set temporal parameters when it distinguished Boldt because it involved different days and different persons.

[7]

However, we need not decide here how we think Lomagro should be interpreted. We are called upon to decide if trial counsel was deficient for failing to raise the issue. If, as this court recognizes, Lomagro can be reasonably analyzed in two different ways, then the law has not been settled. Counsel is not required to object and argue a point of law that is unsettled. Although it might have been ideal for counsel to so object and assert an interpretation of Lomagro that would benefit his client, the fact is that he was not deficient in failing to do so. We conclude that, while a future court may well hold a count like the one at issue here to be duplicitous, the area is murky enough that counsel was not deficient for failing to raise the issue.

We think ineffective assistance of counsel cases should be limited to situations where the law or duty is clear such that reasonable counsel should know enough to raise the issue.

RIGHT TO BE PRESENT WHEN THE TRIAL COURT

COMMUNICATED WITH JURORS AND COUNSEL IN MC MAHON'S ABSENCE

On six occasions following the delivery of the charge to the jury, the jury and the court had communications concerning the conduct of the deliberations. All communications with the jury were recorded. On the first occasion, the jury wanted a copy of the "individual counts as charged and read." The trial court held a conference in chambers. Present in person was the State. McMahon's attorney appeared by telephone. The record does not reflect that the defendant was present. The attorneys and the court arrived at an answer to the jury's request, stating that the information could not be sent into the jury room. The trial court delivered this answer to the jury. The record does not indicate whether that answer was given in open court or if the trial court entered the jury room to give the answer.

On the second occasion, the jurors wrote a note saying that there was much confusion about how long McMahon was in jail and it wanted to know when he was released. It appears that the conference was held in the same manner as the prior conference. After counsel and the court agreed upon an answer, the court went into the jury room to give the answer. Neither counsel nor McMahon was present.

The third communication, at about 11:10 p.m., consisted of a note by the jury that it was "stuck" and wanted to know what to do. The record discloses that the State appeared in person and McMahon's counsel by telephone. Again, McMahon was not present. After conferring with counsel, the trial court decided to let the jury retire for the night and resume deliberations the next day. When the jury was brought into the courtroom to be informed of this, McMahon was not present. Neither was his attorney. Nor does the record reflect whether the State was present in the courtroom.

The fourth time the court and jury communicated was the next day in open court. Both counsel were present, as was McMahon. The trial court inquired of the jury about its deliberative process up to this point, an issue that will be addressed below.

On the fifth occasion, the jury asked whether it was possible for McMahon to have received telephone calls while in jail. It appears that the trial court did not confer with counsel, but simply answered the jury by saying that since this information was not part of the testimony, the jury may not have an answer to that question.

On the sixth occasion, the jury requested a rereading of the information. For the first time, the trial court sought and obtained a waiver from McMahon's counsel that McMahon did not have to be present for this reading. This took place during a telephone conference between the trial court and McMahon's counsel. It seems from the record that the State appeared in person. It does not appear that McMahon's counsel or McMahon himself was present when the jury was brought into the courtroom for the reading of the information. Several hours later, the jury reached its verdict.

[8, 9]

Without question, a criminal defendant is entitled to be present at the trial and to have counsel at every stage where he or she needs aid in dealing with legal problems. See Rogers v. United States, 422 U.S. 35, 38 (1975). A trial court's comments to the deliberating jury without the defendant and his or her counsel being present (unless the defendant has waived that right) deny the defendant his constitutional right to be present at trial. State v. Burton, 112 Wis. 2d 560, 565, 334 N.W.2d 263, 265 (1983).

Our thorough review of the record shows that of the six occasions where the trial court communicated with the jury, only once was there a waiver of defendant's right to be present and that waiver came from McMahon's attorney. Further, the Burton court underscored its "strong, consistently expressed disapproval of a judge's communicating with the jury after it has retired for deliberations, unless the communication is in open court." Id. at 569, 334 N.W.2d at 267. The court wrote:

[W]e do not condone the practice of a judge entering the jury room or communicating with a jury outside of the presence of the defendant and of counsel for the defendant both and the state, even when the judge scrupulously takes a court reporter with him or her to the jury room to record the comments. The judge is a figure of authority and respect during the trial; his or her intrusions into the sanctity of jury deliberations may affect those deliberations. Even a transcript of the judge's communication cannot reveal a judge's facial expressions or tone of voice. Defense counsel and defendant must be present to have the opportunity to observe the judge's demeanor first-hand, to object to statements or request curative statements in the event that the communication may be improper in any way. Id. Here there was waiver for only one of the occasions where the court communicated with the jury. Further, the communications were mostly conducted in the jury room rather than open court. The record demonstrates error under Burton.

[10]

As the Burton court pointed out, however, such error does not automatically entitle the defendant to a new trial; the error may be found to be harmless beyond a reasonable doubt. Id. at 565, 334 N.W.2d at 265-66. We find harmless error here for the following reasons. First, the trial court's excursions into the jury room were preceded on all but one occasion by a telephone conference between attorneys where every party to the conversation agreed to what the trial court should tell the jury. So, those forays by the trial court were not the product of some unilateral decision by the trial court about what to say to the jury. Second, none of the communications suggested anything of such substantive nature that defendant's presence could have aided in dealing with a legal problem. Third, in only one instance was the jury brought into open court without the defendant being present. This was at 11:30 p.m., after the jury had deliberated several hours and the jury was told that it could retire for the evening. The sum and substance of the communication was that everyone was tired and everyone should go home.

McMahon claims that the jury's communications about his jail status provide the proof for why the error is not harmless. McMahon was incarcerated in the county jail at the time of the trial. McMahon asserts that his absence could well have inferred to the jury that he was an inmate and, therefore, a person of bad character.

We do not agree. The jury's questions about McMahon's jail status had to do with his ability to effectuate Nannette's recantation. The jury obviously wanted to know if McMahon was in jail at the time of the recantation and, if so, whether he had the ability to call Nannette. There is nothing in the record to suggest that the jury knew or even had an inkling that McMahon was still in jail when the trial took place. We cannot find anything in the record, nor has any part of the record been cited to us, supporting an inference that the jury would have presumed McMahon to be in jail if he was not present during the trial court's prepared answers to the jury's questions.

[11]

While we do not reverse on this ground, we underscore the same theme announced by the Burton court. Communications with the jury should be conducted in open court, not in the jury room. Further, while it is obvious that practical considerations may sometimes obviate the ideal setting where the State, defendant's counsel and defendant are present, defendant's waiver should be clearly expressed on the record. Further, before communications with the jury take place, the court should state, for the record, the people who are present.

THE TRIAL COURT'S INQUIRY INTO THE NUMERICAL DIVISION OF THE JURY DURING ITS DELIBERATIONS

When the jury had deliberated for several hours and sent a question to the trial court at about 11:10 p.m. saying "we are stuck. Now what?", the trial court eventually dismissed the jury for the evening. The next morning, in open court, the trial court asked for the identity of the foreperson. Once identified, the trial court asked the foreperson the following question: "Without asking you who on one side by name or by their votes, conviction or acquittal or whatever, could you just give me two numbers that add up to twelve that express the votes? That would be on your last votes?" The foreperson answered: "Nine guilty and three not guilty—I'm sorry, what did I do wrong?" Another juror remarked, "She just wanted numbers." The trial court then remarked:

All right, okay, this is the situation where I will just strike that we heard the rest, because that is not your vote until your vote—it is not the vote of the jury and the question was just done on the last ballot. The reason I ask that you give me that was so I would have an idea that when you say to me, Judge, we are stuck, are you saying to me, Judge we think we are deadlocked. And then I would have to make that assessment in managing this activity, so nine three at this stage, and when you went into deliberation at two o'clock in the afternoon and stopped at about eleven ten, something like that, it means that I need to ask you one more question. And the question is this. Did that vote change during the course of the day?

The foreperson answered "yes." Subsequently, the trial court said that although it could not send a copy of the information into the jury room, it could reread the information and proceeded to do so. It then gave what is known as the Allen charge and also paraphrased it in the trial court's own words for the jury.

McMahon submits that the trial court's inquiry into the numerical division of the jury's deliberations constitutes reversible error "in and of itself," and asks this court to adopt a per se rule. Alternatively, McMahon asserts that the inquiry, added to the court's other inquiries, the multiple readings of the instruction asking the jury to agree and its attendant rereading of the information, constituted coercion towards a finding of guilt and necessitates reversal of the convictions.

There is no question that the trial court erred in asking for the numerical division of the jury. The practice has long been frowned upon in Wisconsin. In Mead v. City of Richland Center, 237 Wis. 537, 542, 297 N.W. 419, 422 (1941), our supreme court quoted approvingly from the United States Supreme Court's opinion in Burton v. United States, 196 U.S. 283, 307 (1905), as follows:

"A practice ought not to grow up of [a judge] inquiring of a jury, when brought into court because unable to agree, how the jury is divided; not meaning by such question, how many stand for conviction or how many stand for acquittal, but meaning the proportion of the division . . . . Such a practice is not to be commended, because we cannot see how it may be material for the court to understand the proportion of division of opinion among the jury."

So far as we know, this view still prevails today. We think it makes just as good sense now as it did fifty years ago. There are those who may argue that the more a jury is weighted in one direction, the greater the chance for a unanimous verdict; conversely, the more evenly divided a jury is, the less chance of unanimity. We disagree. No one knows or should know what is going on inside the jury room. While there may be a tiny minority in one case and a sizable minority in another, the tiny minority may be set in their belief while the larger minority may be "soft" votes. We conclude that our supreme court's approval of language labeling the numerical division as immaterial is logically sound. Even if we did not, as an error correcting court, we are bound by our supreme court case law.

The United States Supreme Court not only has considered the practice to be immaterial, but coercive to the point that it has adopted a per se rule prohibiting it. That court wrote:

We deem it essential to the fair and impartial conduct of the trial, that the inquiry itself should be regarded as ground for reversal. Such procedure serves no useful purpose that cannot be attained by questions not requiring the jury to reveal the nature or extent of its division. Its effect upon a divided jury will often depend upon circumstances which cannot properly be known to the trial judge or to the appellate courts and may vary widely in different situations, but in general its tendency is coercive. It can rarely be resorted to without bringing to bear in some degree, serious although not measurable, an improper influence upon the jury, from whose deliberations every consideration other than that of the evidence and the law as expounded in a proper charge, should be excluded. Such a practice, which is never useful and is generally harmful, is not to be sanctioned.

The failure of ... counsel to particularize an exception to the court's inquiry does not preclude this court from correcting the error. This is especially the case where the error, as here, affects the proper relations of the court to the jury, and cannot be effectively remedied by modification of the judge's charge after the harm has been done.

Brasfield v. United States, 272 U.S. 448, 450 (1926) (citations omitted). Thus, the United States Supreme Court has declared it to be a per se rule that federal jurisdiction trial courts may not ask about the numerical division of a deliberating jury and that failure to object is not an important or useful prerequisite.

The State correctly notes that Brasfield was not based on constitutional interpretation, but upon the United States Supreme Court's supervisory power over the federal judiciary and is therefore not controlling. It further notes that while Mead condemns the practice, our supreme court did not rule it to be a per se error. The State argues that we should hold McMahon to waiver since the question is not of constitutional dimension.

[12]

Waiver is not a jurisdictional defect, but one of administration. See Terpstra v. Soiltest, Inc., 63 Wis. 2d 585, 593, 218 N.W.2d 129, 133 (1974). The reason why we view the waiver rule with favor is because failure to bring a matter to the trial court's attention denies the trial court an opportunity to rule on the matter after consideration. Notice allows the trial court to prevent error from occurring. When no motion or objection is interposed, it is difficult for the appellate court to say that a trial court "erred" when it was never given the opportunity to rule on the matter in the first place. See id.

However, we can reach the question in the proper exercise of our discretion despite the fact that there was no objection. See id. We do so in this case for the same reason expounded by the United States Supreme Court in Brasfield. It is the trial court that suddenly and unilaterally asked for the numerical division. There was little or no time to object. The question before us affects the proper relations of the jury and could not have been effectively remedied by modification of the judge's question after it was asked. We will reach the merits.

The State asks that we not adopt the per se rule of the United States Supreme Court. McMahon asks that we do. We will not; indeed, we cannot. Such a rule would be in the nature of a supervisory order and this court has no power to issue such orders. If a per se rule is going to be adopted, it will have to be from our supreme court.

So the question remaining is whether the error was harmless. The State suggests that it was harmless because there was no affirmative evidence of coercion. The State points out that the trial court did not ask the jury foreperson to identify which jurors favored conviction and which favored acquittal. The trial court did not ask about the specific counts in the information. The trial court did not implicitly or explicitly suggest which way the jury should vote. The State further notes that, while the trial court described different ways in which the jury might reach a verdict, it did not order the jurors to come back with a verdict. The trial court told the jury to "look at all the counts that you can agree to and work as best you can and as hard as you can." The trial court also complimented the jurors on their work and told them that their needs would be looked after during deliberations.

The State further asks that we look to the very existence of three acquittals as proof that the jury was not coerced. In the State's view, the split verdicts suggest that the jury felt no pressure to reach a particular verdict. The fact that it deliberated for many hours after the trial court's communication also proves that the jury felt no coercion. The State also contends that the jury's call for additional information after the communication is further proof of the lack of coercion. The State argues that this is not a case "where a draconian circuit judge browbeat a jury into rendering verdicts."

[13]

We are disposed to agree with the State. On the basis of the record before us, nothing the trial court did after finding out the numerical division can be said to have been coercive. Of course, that is the problem. As the Brasfield Court pointed out, the effect on the jury of having requested the numerical division cannot be known to us. That is why the Court fashioned a per se rule. We do not know what the effect of the trial court's question was on this jury. Nor can we speculate. We can only look at the totality of circumstances to make our determination. See Richardson v. State, 508 So.2d 289, 293 (Ala. Crim. App. 1987) ("A totality of the circumstances or a case-by-case approach thus best serves the underlying purpose here: to protect the jury deliberation process from undue influence or coercion.").

This case is not like Mead. There, the judge asked the numerical division and was told that it was eight to four. After hearing the division, the judge told the minority jurors to consider whether they were "warranted in standing on their views as against that of their fellow jurors." Mead, 237 Wis. at 540, 297 N.W. at 420-21. The trial court further said the "four who do not agree ... might well consider whether their judgment is better than the eight." Id. at 539, 297 N.W. at 420. The trial court also alluded to the building getting cold and that it was becoming uncomfortable. The supreme court found these comments objectionable.

[14]

There is nothing in the record here showing that the trial court spotlighted or singled out the three minority jurors. While the three jurors may have felt singled out and made uncomfortable anyway simply due to the trial court's asking about the numerical division we have no way of knowing. In the absence of a per se rule, we cannot reverse on this ground.

POLLING THE JURY

[15]

McMahon's counsel did not ask for the jury to be polled when it came back with split verdicts. McMahon argues ineffective assistance of counsel. We have already described the standard for our review of ineffective assistance of counsel questions. We see no ineffective assistance here. Simply because a jury renders a split verdict does not mean that one or more of the jurors was uncertain of the verdict. Without some showing of uncertainty in the record, we do not believe that error ensued by failing to ask for a polling of the jury.

OTHER ACTS EVIDENCE

[16]

McMahon argues that Nannette's original accusation, offered by the State as a prior inconsistent statement, was filled with other acts evidence which prejudiced him. There was no objection to the statement on this basis and the issue is waived. Alternatively, McMahon raises the issue anew under the guise of an ineffective assistance of counsel argument. We agree with the State that there are at least three reasons why the claim fails. First, the allegations were relevant and admissible to impeach Nannette's recantation and her claim that no sexual contact occurred between herself and McMahon. See State v. Williamson, 84 Wis. 2d 370, 382-83, 267 N.W.2d 337, 343 (1978), overruled on other grounds by Manson v. State, 101 Wis. 2d 413, 422, 304 N.W.2d 729, 734 (1981). Second, in impeaching, the State also was able to show that McMahon was sexually inclined toward Nannette—which goes to motive. Third, the allegations were relevant and admissible to furnish the context of the charged crimes and the relationship between McMahon and Nannette. State v. Shillicut, 116 Wis. 2d 227, 236-37, 341 N.W.2d 716, 720 (Ct. App. 1983), aff'd, 119 Wis. 2d 788, 350 N.W.2d 686 (1984).

OTHER INEFFECTIVE ASSISTANCE OF COUNSEL CLAIMS

McMahon raises a host of other ineffective assistance claims—all of which are without merit. We will address them in turn. First, he faults trial counsel for failing to challenge the time periods in count four. The event surrounding count four concerned the time between October 1988 and May 1989 when Nannette was visiting. Waukesha while living at a foster home in Oconomowoc. Nannette stated that a pornographic videotape was watched on this occasion. McMahon argues that his ability to offer an alibi defense was impaired. He has not shown us how his ability was impaired. Frankly, this issue is inadequately briefed and we decline to address it further.

Regarding count five, Nannette claimed that an incident occurred in the shower around the time that McMahon had an eviction action pending. McMahon notes that the time frame was a period of two years and one week, but the charge purported to be a one-week period in 1989. McMahon claims that trial counsel could easily have narrowed the time frame, but made the most meager of efforts to do so or defend against it. We see no deficient performance. Trial counsel had obtained Nannette's recantation of the entire event. This coupled with McMahon's testimony surrounding his memory of the eviction was an adequate defense to the charge.

Next, he faults trial counsel's failure to have the voir dire and closing arguments recorded. He cites no authority to say that this is ineffective assistance. He also concedes that trial counsel preserved an objection to the prosecutor's closing arguments by making an appropriate record. We will not further entertain the question.

Next, McMahon complains that trial counsel failed to adequately prepare him for trial. The postconviction testimony, however, is that trial counsel met with McMahon six to ten times before trial and discussed the prospective testimony. There is no showing of how any additional preparation would have helped. We see no error.

Next, McMahon asserts that trial counsel failed to object to testimony that he was incarcerated before and during trial. We have found nothing suggesting that the jury knew McMahon was incarcerated during trial. Nor has this been cited to us by McMahon. Regarding testimony that he was at one time incarcerated, the only possible effect this testimony could have had would be to answer the question of whether McMahon could have pressured Nannette to recant if he was in jail. In our view, the fact that the jury may have thought McMahon to be in jail and, therefore, less able to influence Nannette would have bolstered Nannette's recantation. Such an inference would not have hurt McMahon; it would have helped.

Next, McMahon faults a variety of trial counsel's trial strategies. We do not choose to dwell at length upon all these allegations because we believe it sufficient to state that all of trial counsel's strategic choices were professionally reasonable and were not prejudicial.

Next, McMahon claims that trial counsel failed to adequately rebut the expert witness. We conclude that he cross-examined the expert in a professionally reasonable manner. He received an admission that no two individuals respond the same way to traumatic experiences and that most individuals who have been sexually assaulted do not recant. There was a defense basis created to discount the witness' testimony.

[17]

Finally, McMahon argues that counsel failed to present certain evidence. Our confidence in the reliability of this verdict is not hampered on this ground.

INTERESTS OF JUSTICE

We are not convinced that this case should be reversed in the interests of justice.

By the Court.—Judgment and order affirmed in part, reversed in part and cause remanded with directions.

NOTES

[†] Petition to review denied.