State v. Kienitz

221 Wis. 2d 275 (1998) 585 N.W.2d 609

IN RE the COMMITMENT OF Peter KIENITZ:
STATE of Wisconsin, Plaintiff-Respondent,
v.
Peter KIENITZ, Defendant-Appellant.^[†]

No. 97-1460.

Court of Appeals of Wisconsin.

Submitted on briefs April 7, 1998. Decided July 30, 1998.

On behalf of the defendant-appellant, the cause was submitted on the briefs of Suzanne Hagopian, assistant state public defender, of Madison.

On behalf of the plaintiff-respondent, the cause was submitted on the brief of James E. Doyle, attorney general, and Mary E. Burke, assistant attorney general.

Before Eich, C.J., Vergeront and Roggensack, JJ.

VERGERONT, J.

Peter Kienitz appeals from an order committing him to a secure mental health facility as a sexually violent person under Chapter 980, STATS. After a trial to the court, the court determined that Kienitz had been convicted of a sexually violent offense and is dangerous because he suffers from a mental disorder, pedophilia, which makes it substantially probable that he will engage in acts of sexual violence.^[1] Kienitz contends that "substantially probable" means an "extreme likelihood" and the evidence does not support the trial court's determination that it is substantially probable beyond a reasonable doubt that he will engage in acts of sexual violence. If the evidence is sufficient, Kienitz contends, the statute is unconstitutional as applied to him because it is too imprecise to satisfy due process. Finally, Kienitz argues that the State denied him his Fifth Amendment right to remain silent by presenting testimony that he refused to be interviewed by the State's experts.

We conclude that "substantially probable" means "considerably more likely to occur than not to occur," and that the evidence is sufficient to support the trial court's determination. We also conclude there is no merit to Kienitz's constitutional challenge. We do not address the Fifth Amendment issue because we conclude that Kienitz waived this issue by not raising it in the trial court. We therefore affirm.

BACKGROUND^[2]

The evidence of Kienitz's criminal history shows he was convicted of indecent behavior with a child in 1963, and sentenced to ten-years' probation with psychiatric treatment. In 1966, he was found in violation of the terms of probation because he was molesting young boys and was confined in the Wisconsin State Prison for an indefinite term. He was discharged in 1973. In November 1977, he was convicted of firstdegree sexual assault with boys and was sentenced in February 1978 to a commitment to the Department of Health and Social Services under § 975.06, STATS., and five-years' probation. His probation was revoked in 1980 when he was arrested on charges of sexually assaulting two boys under the age of thirteen. He was convicted of second-degree sexual assault on those charges.^[3] He was sentenced to an indeterminate term in prison, not to exceed eight years, and was confined in the Mendota Mental Health Institute most of that time. He was conditionally released from Mendota in March 1988, and was supervised by Probation Agent Sandra Reno.

Reno testified at trial as follows. Among other rules of Kienitz's supervision, he was to have no contact with minors, was to report any such incidental contact immediately and was to report all employment to her. After a few months, she learned from another source that he had a job at a bicycle shop that was frequented by children. In the past, Kienitz had asked parents and children to accompany him on bicycle trips, and, after gaining the parents' confidence, would try to see the children alone. The sexual assault for which he was convicted in 1977 occurred when he lured a child to his house saying he was going to help repair the child's bicycle. A search conducted of his house in 1988 turned up a knife, which violated a rule against possessing a knife; a picture of parents and their two boys whom he accompanied on a bicycle trip; another roll of film he had taken of a very young child; and a list of names and addresses and phone numbers, some of children, including a fifteen-year-old boy whom Kienitz had invited to his house on three occasions to look at computer equipment and bicycles he was repairing. Six pieces of rope were found in a backpack in Kienitz's car, and that was considered significant because in the 1963 offense and in two of the three offenses that occurred between 1979 and 1980, he tied the boys up. Based on the physical evidence and the information discovered in this search, Kienitz's supervision was revoked on April 16, 1988. Kienitz was returned to Mendota with a mandatory release date of October 4, 1995.

Sandy Collins, a psychiatric nurse at Mendota, testified to Kienitz's angry and uncooperative behavior toward other patients and to the staff. She also described two incidents that occurred in December 1994: Kienitz received in the mail material considered pornographic depicting minors and, on another date, a magazine called "Family Fun," a craft magazine with many pictures of young children, a few of which might be considered provocative to a person with a sexual attraction to children. In another incident in May 1995, Kienitz attempted to send out some computer diskettes that the labels indicated were for use by the recipient's two children. When the staff questioned him on the contents, he demanded the diskettes and they were given back to him.

The State presented two expert witnesses, Donald Irwin, Ph.D., director of psychology at the Winnebago Mental Health Institution, and Ronald Sindberg, Ph.D., a psychologist at Mendota. The defense presented Michael Caldwell, Ph.D., a psychologist also employed at Mendota. All three agreed that Kienitz had the mental disorder of pedophilia. The State's experts opined that this mental disorder creates a substantial probability that Kienitz will engage in future acts of sexual violence, but Dr. Caldwell disagreed, testifying that there was only a fifty-fifty chance that Kienitz would do so. The experts reviewed Kienitz's correctional, psychiatric and institutional records, but only Dr. Caldwell interviewed Kienitz because Kienitz did not wish to speak to the other two.

Dr. Irwin described the methodology he used to form his opinion of the probability that Kienitz will engage in future acts of sexual violence. He reviewed research literature on predicting sexual violence and recidivism of sexual offenders and applied the factors predictive of recidivism from the research actuarial data to Kienitz, based on his review of Kienitz's records. Dr. Irwin explained that he made a clinical judgment as to which factors were the most important, because not all the factors were equally strong as predictors. Dr. Irwin testified that he weighed the predictive factors Kienitz possessed against those he did not possess and, in Dr. Irwin's opinion, the former far outweighed the latter. By "substantial probability," Dr. Irwin testified he meant "greater than more likely than not."

Dr. Irwin testified that, in addition to pedophilia, Kienitz was diagnosed with a personality disorder, NOS (not otherwise specified). He viewed Kienitz's personality disorder as significant because, in Dr. Irwin's opinion, pedophilia was a disorder that could not be cured but could be controlled if the person was willing to acknowledge the addiction and learn to control it. Kienitz was unwilling to do that because of his personality disorder, which was characterized by a person being a loner, manipulative, unwilling to conform to social expectations, angry and provocative.

Dr. Irwin also used the violence risk assessment guide (VRAG), which was also used by Dr. Caldwell, Kienitz's expert. The VRAG predicts recidivism for sex crimes as well as violent non-sex crimes. Dr. Irwin determined that under the VRAG, Kienitz was in the range of individuals who had a recidivism rate of 44% in a seven-year follow-up period and 58% in a ten-year follow-up period. In Dr. Irwin's methodology, the VRAG score was integrated with the clinical information to arrive at a professional opinion. Dr. Irwin testified that, following the recommended procedure for VRAG, he would increase the probability percentage of the VRAG score for Kienitz by at least 10%.

Dr. Sindberg used a methodology similar to Dr. Irwin's. He reviewed the research literature on the risk factors predictive of future sexually violent acts, determined which applied to Kienitz, and then considered Kienitz's participation in treatment. Dr. Sindberg acknowledged that there were some errors in an article compiling risk factors that he reviewed. He described his methodology as using actuarial data as an "anchor," and then using clinical judgment to add to that or alter that. He testified that no test or studies had been done to verify the accuracy of his method.

Dr. Sindberg testified that Kienitz had sixteen out of thirty-one risk factors. Based on the records he reviewed, Dr. Sindberg testified that Kienitz refused and avoided individual therapy and therapy groups related to sexual offenses, but participated to some extent in other therapy. In Dr. Sindberg's view, Kienitz had shown little, if any, treatment progress relevant to sexual offenses that might counteract the danger suggested by the presence of the risk factors. However, on cross-examination, Dr. Sindberg acknowledged that Kienitz participated in treatment programs that he had not known about, some appropriate for treating pedophilia.

Dr. Caldwell's testimony criticized the methodology used by Dr. Irwin and Dr. Sindberg to predict Kienitz's probability for future acts of sexual violence. He testified that in Dr. Irwin's and Dr. Sindberg's methodology there is no formal process for weighing the various factors correlated with recidivism and for determining the overlap among the factors and their interactions with each other; rather, they rely on clinical judgment for this. According to Dr. Caldwell, studies have shown that predictions of recidivism based on clinical judgments are accurate not more than 50% or 52% of the time. In the actuarial approach, which Dr. Caldwell testified is more accurate than the clinical approach in predicting risk, factors that have been found to correlate with reoffense are weighed or placed in a formula to either predict a cutoff score or to generate a probability estimate. In the actuarial approach, clinical judgment does not enter into how the factors should be interpreted or applied. Dr. Caldwell identified five actuarial studies that looked retrospectively at a population of offenders, and he identified the accuracy of each study in predicting the reoffenders. He then determined whether Kienitz would have been predicted to reoffend under each study.

VRAG was one of the five studies, and Dr. Caldwell believed that to be the most accurate. Using VRAG, he found a seven-year probability of 35% and a ten-year probability of 48%. Since Kienitz was now fifty-eight years old, Dr. Caldwell opined that the probability of reoffense after ten years would increase only a little over 48%. He explained in detail his disagreement with the process by which Dr. Irwin arrived at his VRAG score.

The other four studies Dr. Caldwell used generated a cutoff score for predicting reoffense, unlike VRAG, which predicts probability. Two predicted Kienitz to reoffend and two did not. Dr. Caldwell viewed these mixed results as consistent with the VRAG result of 48% over ten years. In Dr. Caldwell's opinion, Kienitz's refusal of treatment and his failure to respond to treatment were insignificant in predicting his recidivism because most of the treatments offered were not proven to be effective in reducing recidivism. He did not view Kienitz's provocative behavior in the institution as relevant to the likelihood of his reoffending because there is not a "straight line connection" between institutional adjustment and recidivism, and because persons with Kienitz's passive aggressive personality traits generally do not do well in a controlled setting.

The trial court issued detailed findings of fact and conclusions of law and a memorandum opinion, concluding that Kienitz suffers from a mental disorder, pedophilia, which makes it substantially probable that he will engage in future acts of sexual violence. The memorandum decision thoroughly summarized each of the three experts' opinions. The court found Dr. Caldwell to be more persuasive than Dr. Irwin and Dr. Sindberg because the methods of the latter two were not validated by any scientific testing and because they used clinical judgment, which, the court found, was shown to be a poor predictor of recidivism. The court found Dr. Caldwell's training and experience in the areas of sex offenders and risk assessment to be more extensive then those of the other two psychologists. The court also stated that "the testimony of the experts in this case has been useful and informative and I have relied heavily on it."

The court stated that Dr. Caldwell's testimony, standing alone, would not support a determination that future acts of sexual violence were substantially probable. However, in the court's view, Reno's testimony of her supervision of Kienitz in 1988 strengthened the likelihood that Kienitz would engage in future violence:

At that time it had been 25 years since [Kienitz's] first conviction for a sex offense. He had been imprisoned from 1966 until 1973, and again from 1980 to 1988. If incarceration was to have an effect on his behavior, it should have occurred by then. He had been on probation twice previously and had been revoked both times. Again, if supervision in the community was to improve Mr. Kienitz's behavior, the improvement should have taken place by that time. While on parole in 1988, Mr. Kienitz was 50 years old. He was on intensive supervision with rules designed to help him avoid contact with children and the opportunity to re-offend. He broke these rules in many, serious respects. The violations were very deliberate....

It is now 8 years later. Mr. Kienitz has engaged in no significant treatment for his pedophilia since being reincarcerated in 1988. He does not communicate any plan to deal with his disease. If released, he will not be under supervision. Experience over the last 33 years of his life, especially the most recent episode of being in the community, coupled with the test results obtained by Dr. Caldwell, persuade me beyond a reasonable doubt that there is a substantial probability that Mr. Kienitz will engage in future acts of sexual violence if released.

Kienitz filed post-verdict motions that the trial court denied.

DEFINITION OF SUBSTANTIALLY PROBABLE

Before addressing Kienitz's challenge to the sufficiency of the evidence to support a determination that his pedophilia makes it substantially probable that he will engage in future acts of sexual violence, we must decide the meaning of "substantially probable." During closing argument, the trial court asked the prosecutor what he thought this meant, and the prosecutor answered: "I agree with Dr. Irwin that it is more than more than likely." Defense counsel argued in closing that "substantially probable " means more than "more likely than not," because if the legislature meant the latter, it would have said so. Defense counsel told the trial court that "substantially probable" means "you have to really get up in the certainty range to at least the two-thirds area where you are twice as likely than not to be correct,... anything less would be unconstitutional as applied." He also argued that it means a "strong probability," "much over 50%."

The trial court did not discuss the meaning of "substantially probable" in its memorandum decision, findings of fact, or conclusions of law, other than to summarize Dr. Irwin's testimony on what he meant by the term. Our understanding of the court's decision on this point is that it perceived the parties as not really disagreeing on the meaning of "substantially probable," —the real dispute as reflected in the closing argument was whether the evidence met the standard. We understand the court to have used as a definition of "substantially probable" "something more than more likely than not." That is consistent with what both parties argued, and we have no doubt that if the court had perceived a dispute between the parties on the meaning of this term, it would have addressed that dispute and resolved it, since the court's findings, conclusions and memorandum decision are thorough.

On appeal each party has shifted its position from that argued before the trial court. Kienitz argues that "substantially probable" means "extreme likelihood," while the State contends that it means "likely." The State also argues that "substantially probable" is a term that requires no definition, citing State v. Zanelli, 212 Wis. 2d 358, 374-76, 569 N.W.2d 301, 308 (Ct. App. 1997).^[4] In Zanelli, we decided that the trial court's failure to define the term "substantially probable" in its jury instructions in a Chapter 980 trial did not violate the defendant's right to due process on the ground of vagueness. Id. at 375-76, 569 N.W.2d at 308.^[5] It does not follow, however, that we may not or ought not define the term when the parties dispute its meaning and a resolution is required to decide a challenge to the sufficiency of the evidence.

[1-3]

The issue presented is one of statutory construction. That is a question of law, which we review de novo. State v. Setagord, 211 Wis. 2d 397, 406, 565 N.W.2d 506, 509 (1997). The purpose of statutory interpretation is to discern the intent of the legislature. Id. To do so, we first consider the language of the statute. If the language of the statute clearly and unambiguously sets forth the legislative intent, we apply that to the case at hand and do not look beyond the statutory language to ascertain its meaning. Id. A statute is ambiguous when it is capable of being understood in two or more different senses by reasonably wellinformed persons. Id. at 406, 565 N.W.2d at 510. However, a statute is not rendered ambiguous merely because the parties disagree as to its meaning. Id. If a statute is ambiguous, we look to the scope, history, context, subject matter and object of the statute in order to ascertain legislative intent; but resort to legislative history is not appropriate in the absence of a conclusion of ambiguity. Id.

[4, 5]

In the absence of a statutory definition, we construe words according to their common and approved usage, which may be established by a dictionary. Swatek v. County of Dane, 192 Wis. 2d 47, 61, 531 N.W.2d 45, 50 (1995). The need to look to a dictionary to establish the usual meaning of words does not mean the statute is ambiguous. State v. Sample, 215 Wis. 2d 486, 498-99, 573 N.W.2d 187, 192 (1998).

[6]

While there are many alternative dictionary definitions of "probable," we conclude that the most common and appropriate definition in this context is "[h]aving more evidence for than against," or "likely." BLACK'S LAW DICTIONARY 1201 (6th ed. 1990). See also WEBSTER'S THIRD INTERNATIONAL DICTIONARY 1806 (1993) ("based on ... fairly convincing though not absolutely conclusive ... evidence," synonymous with "likely"). Similarly, while there are many definitions of "substantial" and "substantially," we conclude that the most common and appropriate definition in this context is "considerable in amount, value or worth." WEBSTER'S THIRD INTERNATIONAL DICTIONARY 2280 (1993). See also AMERICAN HERITAGE COLLEGE DICTIONARY 1354 (3d ed. 1993) ("[c]onsiderable in importance, value, degree, amount, or extent"). Because "probable" on its own ordinarily means "likely," the addition of "substantially" must add to "likely"; otherwise "substantially" is superfluous, a result we avoid in construing statutes. See Frederick v. McCaughtry, 173 Wis. 2d 222, 226, 496 N.W.2d 177, 179 (Ct. App. 1992). We therefore reject the State's position that "substantially probable" means "likely," because in this proposed construction "substantially" has no function. We conclude that "substantially probable" means "considerably more likely to occur than not to occur." We also conclude that this definition is equivalent to the one implicitly used by the trial court.

[7]

We also reject Kienitz's definition of "extreme likelihood" because it equates "substantially" with "extreme" or "extremely." The common usage of "substantially" is not synonymous with any of the possible definitions of "extremely." See WEBSTER'S THIRD INTERNATIONAL DICTIONARY 807 (1993), (providing these definitions, among others, of "extreme": "a very pronounced or excessive degree"; "the utmost conceivable or tolerable degree"; "the utter limit"; "maximum"). "Extreme likelihood" conveys a greater degree of certainty than "substantially probable."

We do not define "substantially probable" with a minimum percentage, as Kienitz argued to the trial court. The legislature chose not to use a percentage in defining the degree of certainty of future sexual violence, and the ordinary meaning of the words the legislature chose do not yield a fixed percentage. The lack of a fixed percentage as part of a definition does not, in our view, make the statute ambiguous.

Because the ordinary meaning of "substantially probable"—"considerably more likely to occur than not to occur"—provides a clear definition of the statutory language, our analysis properly ends here. However, both parties apparently view the statutory language as ambiguous, because they base their appellate arguments on the legislative history of Chapter 980, STATS. In particular, they base their argument on a drafter's note from the Legislative Reference Bureau^[6] indicating that "substantially probable" was substituted for "likely" in order to make Chapter 980 consistent with the standards for mental commitment in § 51.20(1)(a)2.a-d, STATS.^[7] Then, the parties examine different aspects of the legislative history of Chapter 51, STATS., arriving at opposing conclusions on the meaning of "substantially probable" in § 980.01(7), STATS. Assuming for purposes of discussion that it is proper to consider the legislative history of Chapter 980 in construing "substantially probable," we find the drafter's note and the legislative history of Chapter 51 presented to us by the parties to be so inconclusive as to not aid in construction.

Kienitz starts with the decision in Lessard v. Schmidt, 349 F. Supp. 1078 (E.D. Wis. 1972), which held the then current version of Chapter 51 (§§ 51.001-51.05, STATS., 1971) unconstitutional in certain respects. In discussing the degree of dangerousness constitutionally required, the court used the term "extreme likelihood."^[8] Kienitz argues that the term "substantial probability" in §51.20(1)(a)2, STATS., means "extreme likelihood" because the legislature revised the statute in response to Lessard. Kienitz also relies on an attorney general opinion, 71 Op. Att'y Gen. 34 (1982), which opined that a legislative proposal to change "substantial probability" to "reasonable possibility" in § 51.20(1)(a)2 would likely not be constitutional, referring to Lessard,

There are significant gaps in Kienitz's argument. The recodification of Chapter 51, STATS., in response to Lessard, did not use the term "substantial probability" but instead used the term "substantial risk." See Laws of 1975, ch. 430, § 11 and Legislative Reference Bureau Analysis of June 1976, Special Sess. AB7. By the time these legislative changes were made, there had been two other district court decisions in Lessard, each after vacation and remand from the United States Supreme Court, and the judgment finally specifying the declaratory and injunctive relief did not use the term "extreme likelihood."^[9] Neither the term "substantial risk," nor the Legislative Council Notes,^[10] nor the Legislative Reference Bureau Analysis of June 1976 gives any indication that the legislature intended "substantial risk" to mean "extreme likelihood."

The change from "substantial risk" to "substantial probability" in § 51.20(1)(a)2.a-c, STATS., was made by Laws of 1977, ch. 428, §§ 29, 30. This change had nothing to do with Lessard, but occurred because of two Milwaukee court opinions that held the term "very substantial risk" in then § 51.20(1)(a)2.c, STATS., was unconstitutionally vague. See Information Memorandum 78-26, Changes in Mental Health Code (Laws of 1977, ch. 428), July 5, 1978.^[11]

71 Op. Att'y Gen. 34 (1982) adds little in support of Kienitz's argument.^[12] The opinion is focused not on the meaning of "substantial probability" but rather on whether a change from that standard to a "reasonable possibility" would be constitutional. In discussing the constitutional standard, the attorney general's opinion acknowledged that Lessard's precedential value might be diminished because of its "unusual procedural history," but concluded that, even without regard to Lessard, the proposed amendments would not survive a constitutional challenge. 71 Op. Att'y Gen. at 37. The opinion may perhaps be read to assume that the existing "substantial probability" language was an effort to comply with the "extreme likelihood" language of the first Lessard decision, but, as we have seen, the legislative history of Laws of 1975, ch. 430, § 11 and Laws of 1977, ch. 428, §§ 29, 30 does not support that assumption.

The State's argument based on legislative history is equally unpersuasive. The State argues that the drafter's comments on Chapter 980, STATS., shows that "substantial probability" from Chapter 51, STATS., was considered to be just another term meaning the same thing as "likely." That is a reasonable interpretation of the drafter's comment, but not the only one: it is also reasonable to infer that in choosing another term, something different was intended. The drafter's note is brief and unclear on this point. The State also asserts that "substantial probability" in § 51.20(1)(a)2, STATS., means "likely" because it means something more than "substantial risk" and something less than "very substantial probability." We agree that the Information Memorandum to Laws of 1977, ch. 428, §§ 29, 30 establishes that "very substantial probability" is a higher standard than "very substantial risk"; and common sense tells us that "substantial probability" is less than "very substantial probability." However, we fail to see how this shows that "substantial probability" in § 51.20(1)(a)2 means "likely."

In short, the materials and arguments on legislative history presented to us by the parties raise more questions than they answer about the meaning of "substantially probable" in § 980.01(7), STATS. We are satisfied that our construction based on the ordinary meaning of those words is the correct one.

SUFFICIENCY OF THE EVIDENCE

Kienitz contends that the evidence is insufficient to support a determination beyond a reasonable doubt that his pedophilia makes it substantially probable that he will engage in future acts of sexual violence.^[13] We address first the question of our standard of review, which the parties dispute.

[8]

Relying on K.N.K. v. Buhler, 139 Wis. 2d 190, 198, 407 N.W.2d 281, 285 (Ct. App. 1987), Kienitz contends that we should review the court's findings of fact under the "clearly erroneous standard" and then decide de novo whether the facts as found by the trial court meet the statutory standard. The State responds that we should use the standard articulated in criminal cases for reviewing challenges to the verdict based on the sufficiency of the evidence, under which we reverse only if the evidence, viewed in the light most favorable to the verdict, is so insufficient in probative value and force that it can be said as a matter of law that no trier of fact, acting reasonably, could have found guilt beyond a reasonable doubt. See State v. Burkman, 96 Wis. 2d 630, 643, 292 N.W.2d 641, 643 (1980). This is the standard of review in criminal cases whether the trial is to the court or to the jury. State v. Oppermann, 156 Wis. 2d 241, 246, 456 N.W.2d 625, 627 (Ct. App. 1990). Kienitz replies that this standard is inapplicable because Chapter 980, STATS., sets forth a civil commitment procedure, see State v. Carpenter, 197 Wis. 2d 252, 258, 541 N.W.2d 105, 107 (1995), and therefore the K.N.K. standard is more appropriate, since that was a proceeding for a protective placement under § 55.06, STATS.

We conclude that the proper standard of review is that set forth in Burkman for several reasons. First, although Chapter 980, STATS., is a civil proceeding, the rules of evidence in criminal proceedings apply as do all constitutional rights available to a defendant in a criminal proceeding, and the burden of proof is beyond a reasonable doubt. See § 980.05(1m) and (3), STATS.

Second, in K.N.K., we treated each of the statutory elements as findings of fact, reviewing them under the clearly erroneous standard, and we affirmed the trial court's order for protective placement because there was sufficient evidence to support each element under the clearly erroneous standard. K.N.K., 139 Wis. 2d at 198-204, 407 N.W.2d at 285-88. It is difficult to discern what we reviewed de novo, and therefore K.N.K. does not provide guidance for us here.

Third, we do not see the distinction between a criminal and civil proceeding to be a critical one on the question of our degree of deference to the fact-finder's verdict. In a civil trial to a jury, the deference given to the jury's verdict on review with respect to drawing of inferences, credibility of witnesses, weight of evidence, and evaluation of experts' testimony, is similar to that given a jury in a criminal trial. Compare Sumnicht v. Toyota Motor Sales U.S.A., Inc., 121 Wis. 2d 338, 360, 360 N.W.2d 2, 12 (1984), and Brogan v. Indus. Cas. Ins. Co., 132 Wis. 2d 229, 238, 392 N.W.2d 439, 443 (Ct. App. 1986), with State v. Poellinger, 153 Wis. 2d 493, 503-04, 451 N.W.2d 752, 756 (1990), and Schultz v. State, 87 Wis. 2d 167, 173, 274 N.W.2d 614, 617 (1979). We do not see any reason to apply a different standard of review to a sufficiency of the evidence challenge simply because the court, rather than a jury, finds the facts and applies the law (as interpreted by the court in both situations) to those facts.

We turn now to Kienitz's challenge to the sufficiency of the evidence, which turns on his view of the role of expert testimony in a determination that it is "substantially probable that the person will engage in acts of sexual violence." Kienitz contends that because the court found Dr. Caldwell more persuasive than the State's experts, and because the court acknowledges that Dr. Caldwell's testimony alone did not establish that reoffense was substantially probable beyond a reasonable doubt, the evidence found reliable by the court was insufficient to meet that standard. We reject Kienitz's argument because it depends upon two premises with which we disagree.

First, we do not agree that the court did not rely on any expert testimony except that of Dr. Caldwell. It is evident from the court's statements in its memorandum decision and in its oral decision denying the postconviction motion that, while the court relied more on Dr. Caldwell's testimony than on that of the State's experts, it found "the testimony of the experts in this case ... useful and informative and [it] relied heavily upon it."^[14] There was much testimony from all three experts besides their ultimate opinion on whether it was substantially probable that Kienitz would reoffend, including the nature of Kienitz's disorder, the factors that are predictive of reoffending and those that are not, and how each expert evaluated Kienitz in relation to those factors. The fact-finder may accept certain portions of an expert's testimony while rejecting other portions. State v. Owen, 202 Wis. 2d 620, 634, 551 N.W.2d 50, 56 (Ct. App. 1996). Although the trial court found Dr. Caldwell's methodology more reliable than that of the other two experts, it does not follow that the court could not, or did not, credit portions of their testimony in making its determination.

The second premise of Kienitz's argument is that unless the fact-finder relies on an expert opinion that it is substantially probable that a Chapter 980 respondent will reoffend, there is insufficient evidence regardless of other evidence. We disagree, and conclude that there may be sufficient evidence that acts of sexual violence are substantially probable, even though the fact-finder chooses not to rely on an expert opinion to that effect.^[15]

Kienitz relies on a statement of the Supreme Court in Addington v. Texas, 441 U.S. 418, 429 (1979), which held that due process requires a "clear and convincing" standard of proof in a state involuntary commitment proceeding, but does not require the "beyond a reasonable doubt standard." In that context, Addington discussed the difference between the inquiry in a civil commitment proceeding and a criminal prosecution:

There may be factual issues to resolve in a commitment proceeding [as in a criminal prosecution], but the factual aspects represent only the beginning of the inquiry. Whether the individual is mentally ill and dangerous to either himself or others and is in need of confined therapy turns on the meaning of the facts which must be interpreted by expert psychiatrists and psychologists.

Addington, 441 U.S. at 429.

This brief statement, which is not the holding of the case or central to it, describes the general role of expert testimony in commitment proceedings, but falls far short of requiring that a fact-finder rely on an expert's opinion that it is substantially probable that an individual will reoffend in order to make that determination. A later case, Estelle v. Smith, 451 U.S. 454 (1981), has a more extensive discussion of the role of expert testimony in predicting whether one will reoffend and supports our view that Kienitz is reading too much into Addington. In Estelle, the Court held that the admission of psychiatric testimony based on a court-ordered examination of the defendant in the penalty phase of a capital murder trial violated the defendant's constitutional rights. Estelle, 451 U.S. at 471. In explaining that this ruling would not prevent the State from establishing future dangerousness, the court stated:

Moreover, under the Texas capital sentencing procedure, the inquiry necessary for the jury's resolution of the future dangerousness issue is in no sense confined to the province of psychiatric experts. Indeed, some in the psychiatric community are of the view that clinical predictions as to whether a person would or would not commit violent acts in the future are "fundamentally of very low reliability" and that psychiatrists possess no special qualifications for making such forecasts.

Id. at 472 (citations omitted).

Citing from Jurek v. Texas, 428 U.S. 262 (1976), the Court stated:

"[P]rediction of future criminal conduct is an essential element in many of the decisions rendered throughout our criminal justice system."

Estelle, 451 U.S. at 473. The Court concluded:

While in no sense disapproving the use of psychiatric testimony bearing on the issue of future dangerousness, the holding in Jurek was guided by recognition that the inquiry mandated by Texas law does not require resort to medical experts.

Id.

Kienitz also appears to argue that, not only must an expert testify that reoffending is substantially probable, but that opinion must be based on the actuarial methodology used by Dr. Caldwell. In essence, Kienitz argues that the court in this case could not rely on the testimony of the other two experts regarding their prediction because their method is scientifically unreliable. Kienitz contends that, because the statute requires that reoffending be substantially probable, and there was expert testimony that the recidivism rate for sex offenders in Wisconsin is no more than 30%,^[16] only the actuarial method results in the necessary level of precision. We reject this argument.

[9, 10]

The testimony of Dr. Irwin and Dr. Sindberg was admissible under our Wisconsin standard for admitting expert testimony. See State v. Walstad, 119 Wis. 2d 483, 516, 351 N.W.2d 469, 487 (1984) (expert testimony is admissible if it is relevant, the witness is qualified as an expert, and specialized knowledge will assist the trier of fact). Once admitted, the fact-finder decides reliability, see State v. Peters, 192 Wis. 2d 674, 690, 534 N.W.2d 867, 873 (Ct. App. 1995), and resolves conflicts and inconsistencies in expert testimony. See Schultz, 87 Wis. 2d at 173, 274 N.W.2d at 617. Moreover, in discussing the constitutionality of Chapter 980, STATS., our supreme court in State v. Post, 197 Wis. 2d 279, 541 N.W.2d 115 (1995), noted the lack of consensus in the behavioral sciences on various issues relating to sex offenders, id. at 310-11, 541 N.W.2d at 125, and emphasized the need for separation between legal and medical definitions. Id. at 305-06, 541 N.W.2d at 123. Kienitz's argument not only obliterates the distinction between the legal standard and behavioral science standards, but, in essence, requires adherence to one particular behavioral science methodology.

The trial court as fact-finder was free to weigh the experts' testimony when it conflicted and decide which was more reliable; to accept or reject the testimony of any expert, including accepting only parts of an expert's testimony; and to consider all the non-expert testimony in deciding whether it was substantially probable that Kienitz would commit future acts of sexual violence. Granting these functions to the trial court, we conclude that there was sufficient evidence from which a reasonable fact-finder could find beyond a reasonable doubt that it was substantially probable that Kienitz would commit future acts of sexual violence.

[11]

The trial court's findings and memorandum decision reveal that it placed great weight on the fact that Kienitz had reoffended or prepared to reoffend after each release since his original conviction in 1963. The court found particularly significant Kienitz's actions in 1988, after being released only a few months and after a long period of incarceration and treatment efforts, his denial of the need for treatment^[17] and his recent involvement with materials relating to or for children while in the institution. The weight the court attached to past offenses was supported by all the experts' testimonies on the importance of past offenses in predicting recidivism. The weight the court attached to Kienitz's recent denial of his pedophilia and need for treatment was supported by the testimony of Dr. Sindberg and Dr. Irwin. The court was not obligated to accept the weight Dr. Caldwell assigned the various factors in his scoring of the VRAG, nor was it obligated to choose either Dr. Caldwell's or Dr. Irwin's VRAG score and rely solely on that score as a measure of probability.

Kienitz also argues that the trial court erred in finding that he was convicted of first-degree sexual assault on November 15, 1977, and again on February 10, 1978, because both dates concerned the same charge, with the plea entered on the first date and the conviction entered on the second. The State concedes, and we agree, that this finding is an error. No doubt if it had been brought to the attention of the trial court, the court would have corrected the error. The error does not, however, affect our conclusion on the sufficiency of the evidence. The trial court's determination that it was substantially probable that Kienitz would reoffend was not based on the specific number of prior offenses, as the memorandum decision makes clear.

DUE PROCESS

Kienitz argues that if the evidence is sufficient to support a commitment, then the order for commitment violates his right to due process under the Fifth and Fourteenth Amendments to the United States Constitution and Article I, Section 8 of the Wisconsin Constitution because the standard of dangerousness as applied to him is too imprecise. According to Kienitz, because clinical judgment is not particularly accurate in predicting recidivism by sex offenders, a commitment order that is unsupported by an opinion that uses the actuarial method to reach the conclusion that a future act of sexual violence is substantially probable is too imprecise to satisfy due process.

[12]

We conclude there is no merit to Kienitz's constitutional challenge. We assume that Kienitz means "vague" when he used the word "imprecise." A constitutional challenge of vagueness is founded on the "fair notice" requirement of procedural due process. Zanelli, 212 Wis. 2d at 374, 569 N.W.2d at 308. In Zanelli, we held that "substantially probable" is not so vague that it violates procedural due process. Id. at 372-73, 569 N.W.2d at 307.

[13]

Kienitz also briefly mentions "substantive due process" in his argument. However, as he acknowledges, our supreme court held in Post that Chapter 980, STATS., does not violate substantive due process guarantees of the United States and Wisconsin Constitutions. See Post, 197 Wis. 2d at 293-94, 541 N.W.2d at 118. Specifically, the court held that the statute's definition of dangerousness—that an individual "is dangerous because he or she suffers from a mental disorder that makes it substantially probable that the person will engage in acts of sexual violence," § 980.01(7), STATS.—does not violate substantive due process. Id. at 311-13, 541 N.W.2d at 126.

[14]

Apparently Kienitz is attempting to present a constitutional challenge not addressed in either Post or Zanelli. However, he does not explain how a statutory standard that does not violate substantive due process and is not unconstitutionally vague nevertheless violates his rights to due process, when the standard is correctly applied and supported by sufficient evidence, as we have held it was in this case. And he cites no authority for this argument. We therefore do not discuss it further. See State v. Gulrud, 140 Wis. 2d 721, 730, 412 N.W.2d 139, 142-43 (Ct. App. 1987) (reviewing court will not consider undeveloped arguments).

RIGHT TO REMAIN SILENT

Kienitz contends that his Fifth Amendment right to remain silent was violated when Dr. Irwin and Dr. Sindberg testified that they did not interview Kienitz in forming their opinions because he did not want to be interviewed. Kienitz relies on Zanelli, 212 Wis. 2d at 369-72, 569 N.W.2d at 306-07. There we held that Zanelli's Fifth Amendment right to remain silent was violated when the State's expert testified that Zanelli refused to participate in a formal evaluation and the prosecutor in closing argument commented on this refusal. Id. We decided this issue although the State argued that it was waived because defense counsel made no objection when the alleged errors occurred. Id. at 369, 569 N.W.2d at 306. We observed that although an appellate court generally does not decide issues raised for the first time on appeal, this rule of judicial administration does not deprive the court of the power to decide the issue. We also "doubt[ed] there was a waiver [because] Zanelli pursued a claim of the violation of his right to remain silent through a series of pretrial motions," which the trial court denied. Id.

The State contends that Kienitz waived the Fifth Amendment objection, distinguishing the record in this case from that in Zanelli.^[18] Kienitz replies that he raised the Fifth Amendment issue in a motion to dismiss the petition and therefore did not waive it. We agree with the State that Kienitz waived the right to object to Dr. Irwin's and Dr. Sindberg's testimony on Fifth Amendment grounds, and we conclude that the record presents no compelling reasons to review the issue in spite of the waiver. See Maclin v. State, 92 Wis. 2d 323, 329, 284 N.W.2d 661, 664 (1979) (in exceptional circumstances appellate court will review issues not raised in the trial court).

Kienitz filed a motion to dismiss the petition on several grounds, including that the petition did not allege sufficient facts to show probable cause. Kienitz also moved that:

[I]n deciding this probable cause issue, the court strike any language contained in the petition that relates to communications between [Kienitz] and any treatment provider and strike any language relating to [Kienitz's] refusal of treatment, on the grounds that those communications with treatment providers are privileged and it would be unfair to consider [Kienitz's] refusal of treatment, where Kienitz is placed in the position of getting treatment knowing that communications made in the context of such treatment would be used against him at a later time, such as now.

Although Kienitz states the trial court denied his motion, he provides no citation to the record on that point. As far as we can tell, the trial court never ruled on this point in the motion to dismiss because it dismissed the petition on constitutional grounds, which were eventually reversed. See footnote 2. We see no evidence in the record that Kienitz pursued this issue before or during the trial after remand.

During the trial, the prosecutor asked Dr. Irwin whether he had ever seen Kienitz in order to establish that the subject of the petition was in fact the person whose records Dr. Irwin reviewed. In explaining the circumstances in which he saw Kienitz, Dr. Irwin testified that he went to interview him but Kienitz would not talk to him, so Dr. Irwin saw him for only a half a second. Defense counsel did not object to the testimony that Kienitz would not talk to Dr. Irwin.^[19] On cross-examination, in the context of comparing the basis for clinical opinions on future dangerousness in certain studies to the information Dr. Irwin considered, defense counsel asked, "and you didn't talk to Mr. Kienitz, correct?", to which Dr. Irwin answered no. In response to the prosecutor's questions about what information Dr. Sindberg examined in evaluating Kienitz, Dr. Sindberg testified that Kienitz refused to be interviewed by him. Defense counsel did not object to Dr. Sindberg's responses either.

Dr. Caldwell testified that he interviewed Kienitz, in explaining the foundation for his opinion, and he again referred to the interview in answering a question on whether Kienitz benefited from treatment. The prosecutor did not mention that Kienitz had refused to be interviewed by the State's experts in closing argument. The only reference to whether Kienitz was interviewed or not in the court's decision was in the context of summarizing the foundation for each expert's opinion.

In his post-verdict motions, Kienitz renewed the same request we cited above from the motion to dismiss, but made no argument elaborating on it at the hearing. The court denied this in a brief statement, concluding that the petition was adequate to establish probable cause and "it did not violate any rights of the petition to rely on that information [from Kienitz's medical records and Dr. Sindberg's evaluation]." Neither in the post-verdict motion nor at the hearing on the motion did Kienitz's counsel mention the trial testimony of Dr. Sindberg and Dr. Irwin on Kienitz refusing to be interviewed.

Unlike Zanelli, we do not have doubts here that Kienitz waived objection to the testimony on his refusal to be interviewed by the State's experts. The motion to dismiss on the privilege and refusal of treatment ground was not renewed before trial, and neither on its face nor in its post-verdict presentation does it raise the issue Kienitz now raises—that testimony on Kienitz's refusal to be interviewed violates his Fifth Amendment right to remain silent. An objection must be made with sufficient specificity and prominence so that the trial court understands what it is expected to rule on. See State v. Salter, 118 Wis. 2d 67, 79, 346 N.W.2d 318, 324 (Ct. App. 1984). Not only did defense counsel not object to Dr. Irwin's testimony that Kienitz refused to be interviewed, but on cross-examination he elicited the same testimony to make a point challenging the foundation of Dr. Irwin's opinion.

[15]

We are not persuaded on this record that we should consider Kienitz's challenge to the testimony in spite of waiver. The affirmative use by the defense of the lack of an interview with Kienitz,^[20] combined with the absence of any comment by the prosecutor in closing, as well as the context in which the challenged testimony occurred—identification and foundation —convince us that there are no compelling reasons to address this issue.

By the Court.—Order affirmed.

NOTES

[†] Petition to review granted.

[1] Section 980.01(7), STATS., defines sexually violent person as:

(7) "Sexually violent person" means a person who has been convicted of a sexually violent offense, has been adjudicated delinquent for a sexually violent offense, or has been found not guilty of or not responsible for a sexually violent offense by reason of insanity or mental disease, defect or illness, and who is dangerous because he or she suffers from a mental disorder that makes it substantially probable that the person will engage in acts of sexual violence.

[2] The petition filed by the State alleging that Kienitz is a sexually violent person eligible for commitment under ch. 980, STATS., was originally filed on October 2, 1995. The court found probable cause to believe that Kienitz was a sexually violent person and ordered him detained. It then granted a motion to dismiss the petition, concluding that ch. 980 was unconstitutional. The State appealed that ruling to this court, and we stayed Kienitz's release pending appeal. While the appeal was pending, the supreme court decided State v. Carpenter, 197 Wis. 2d 252, 541 N.W.2d 105 (1995), and State v. Post, 197 Wis. 2d 279, 541 N.W.2d 115 (1995). These cases upheld ch. 980 against the constitutional challenges, except that the supreme court held a jury trial was required in recommitment proceedings on equal protection grounds. See Post, 197 Wis. 2d at 328-29, 541 N.W.2d at 132-33. Following Carpenter and Post, we reversed the trial court's dismissal of the petition against Kienitz, and remanded for further proceedings. Kienitz waived his right to a jury trial, and the trial to the court took place over three days in October 1996.

[3] There is no dispute that this conviction was for a "sexually violent offense" within the meaning of § 980.01(6), STATS.

[4] The State also argues that Kienitz waived the right to argue "extremely likely" as a definition, because he did not argue that below. We disagree with the State's contention. The issue concerning the meaning of "substantially probable" was raised in the trial court—in fact, the trial court asked the parties to address it in their closing argument and both parties argued a position on this issue. Therefore, the issue is not being raised for the first time on appeal. As we have noted above, both parties have shifted positions on appeal. We see no reason not to consider the positions each party advances on appeal. They are legal positions, neither party is prejudiced, and it aids in a complete analysis of the issue.

[5] The jury instruction requested in State v. Zanelli, 212 Wis. 2d 358, 375, 569 N.W.2d 301, 308 (Ct. App. 1997), was:

A result has a substantial probability of occurring if there are indications strong enough to alert a reasonably prudent person not only to the possibility of the result occurring, but the indications also must be sufficient to forewarn the person that the result is highly likely to happen.

We stated that although due process did not require an instruction defining "substantially probable," the court had the discretion to give this instruction. Id. at 376,569 N.W.2d at 308.

[6] LRB Drafter's Note to 1994 A.B. 3, (LRB-2975/P2dn) (Oct. 15, 1993), provides in part:

One of Mr. Bucher's comments was that the bill should use words already used in the statutes. I have changed some of the terminology to try to do this. For instance, instead of saying that a person must be "likely" to commit predatory acts of sexual violence, I changed the language to say the person must be "substantially probably" to commit such acts. Cf. s. 51.20(1)(a)2. a to d, STATS. Mr. Bucher also specifically suggested changing "released to be at large" to "dangerous to himself or others" since the latter follows the "conditional discharge law", which I take to mean conditional discharge under s. 971.17(3)(e) and (4), STATS. However, s. 971.17, STATS., does not speak about "dangerousness" in so many words; instead, it requires an inquiry into whether the person poses a "significant risk" of harm to others or their property. I used the "substantially probable" language throughout the redraft for internal consistency and because it is consistent with s. 51.20, STATS. In addition, the "significant risk" language in s. 971.17, STATS., is, I think, a somewhat lesser standard. Do you want to consider using the "significant risk"?

[7] Section 51.20(1)(a)2.a-d, STATS., provides in part:

2. The individual is dangerous because he or she does any of the following:

a. Evidences a substantial probability of physical harm to himself or herself as manifested by evidence of recent threats of or attempts at suicide or serious bodily harm.

b. Evidences a substantial probability of physical harm to other individuals as manifested by evidence of recent homicidal or other violent behavior, or by evidence that others are placed in reasonable fear of violent behavior and serious physical harm to them, as evidenced by a recent overt act, attempt or threat to do serious physical harm....

c. Evidences such impaired judgment, manifested by evidence of a pattern of recent acts or omissions, that there is a substantial probability of physical impairment or injury to himself or herself....

d. Evidences behavior manifested by recent acts or omissions that, due to mental illness, he or she is unable to satisfy basic needs for nourishment, medical care, shelter or safety without prompt and adequate treatment so that a substantial probability exists that death, serious physical injury, serious physical debilitation or serious physical disease will imminently ensue unless the individual receives prompt and adequate treatment for this mental illness.

[8] The context of the Lessard court's use of this term is:

However, [the Supreme Court's] approval [in another case] of a requirement that the potential for doing harm be "great enough to justify such a massive curtailment of liberty" implies a balancing test in which the state must bear the burden of proving that there is an extreme likelihood that if the person is not confined he will do immediate harm to himself or others. Although attempts to predict future conduct are always difficult, and confinement based upon such a prediction must always be viewed with suspicion, we believe civil confinement can be justified in some cases if the proper burden of proof is satisfied and dangerousness is based upon a finding of a recent overt act, attempt or threat to do substantial harm to oneself or another.

Lessard v. Schmidt, 349 F. Supp. 1078, 1093 (E.D. Wis. 1972) (citation omitted).

[9] Lessard v. Schmidt, 349 F. Supp. 1078 (E.D. Wis. 1972), was vacated and remanded, 414 U.S. 473 (1974), because it did not specifically and in reasonable detail specify the acts sought to be restrained. On remand, in Lessard v. Schmidt, 379 F. Supp. 1376 (E.D. Wis. 1974), the court spelled out the due process rights required in civil commitment proceedings and did not use the term "extreme likelihood" in describing the substantive standard but instead ordered that "[t]he findings and standard of proof necessary for an order of commitment are mental illness and imminent dangerousness to self or others beyond a reasonable doubt' based at minimum upon a recent act, attempt, or threat to do substantial harm." Lessard, 379 F. Supp. at 1381. This decision was again vacated and remanded, Schmidt v. Lessard, 421 U.S. 957 (1975), for consideration of a recent Supreme Court case concerning federal intervention in state civil proceedings. Upon remand, the district court decided that its prior grant of relief was proper and it reinstituted the prior judgment of the court. Lessard v. Schmidt, 413 F. Supp. 1318 (E.D. Wis. 1976).

[10] Legislative Council Notes, 1976, Wis. STAT. ANN. § 51.20 (West 1997).

[11] The phrase "very substantial probability" was substituted for "very substantial risk" in that subsection and the phrase "substantial probability" was substituted for "substantial risk" in § 51.20(1)(a)2.a and b, STATS., Laws of 1977, ch. 428, §§ 29, 30. The following year "very" was deleted from then § 51.20(1)(a)2.c. Laws of 1979, ch. 336, § 7.

[12] An attorney general opinion is not binding on the courts, and we give them such persuasive effect as we deem warranted. State v. Gilbert, 115 Wis. 2d 371, 380, 340 N.W.2d 511, 516 (1983).

[13] It is undisputed that Kienitz suffers from pedophilia and that it is a "mental disorder" within the meaning of § 980.01(2), STATS., which defines "mental disorder" as "a congenital or acquired condition affecting the emotional or volitional capacity that predisposes a person to engage in acts of sexual violence."

[14] In denying the post-verdict motion on sufficiency of the evidence, the trial court stated:

I think that [Kienitz's trial counsel] perhaps is struggling with the fact that I chose not to rely particularly heavily on the State's two psychological experts and relied frankly more heavily on his own expert and the testimony of [former parole agent] Ms. Reno. All of the testimony supports the conclusion that I reached.

The court is not bound by the conclusion of any expert. The court can consider the opinions of experts insofar as they assist in reaching a decision, and I did consider the opinions of all the witnesses in reaching my conclusion, but I do believe that the evidence taken as a whole is sufficient to establish that the State met its burden of proof.

[15] Although the parties debate whether there must be expert medical testimony on the issue of future acts of sexual violence, the issue presented in this case is much narrower, because there was indeed expert testimony on that issue.

[16] It appears from the testimony that the rate Kienitz refers to here is based on new convictions rather than conduct for which the perpetrator may not be arrested or convicted.

[17] The court made these findings: In 1990, it was reported that Kienitz did not intend to involve himself in treatment and that he viewed himself as "cured" of pedophilia; in 1994, he stated he was no longer a pedophile and declined treatment; in 1994, prior to the filing of the petition, he refused to participate in any planning for community living and expressed a preference to remain institutionalized until his mandatory release date rather than have the restrictions of parole and he denied the need for treatment as recently as August 1996.

[18] The State makes two additional arguments: that our Fifth Amendment analysis in State v. Zanelli, 212 Wis. 2d 358, 369-72, 569 N.W.2d 310, 306-07 (Ct. App. 1997), is flawed, and that any error is harmless, an issue which Zanelli did not consider. We do not address these arguments because of our decision on waiver.

[19] When Dr. Irwin testified he could not identify Kienitz based on his half second view of him, the prosecutor wanted defense counsel to stipulate that Kienitz was the person referred to in the report. Defense counsel objected to a stipulation, and the court informed the prosecutor that it would assume Dr. Irwin reviewed Kienitz's records and not someone else's, since the defense was not asserting otherwise.

[20] In State v. Keith, 216 Wis. 2d 61, 82, 573 N.W.2d 888, 897 (Ct. App. 1997), we held that when the defense counsel questioned, in opening statement and through cross-examination, the quality of the State's expert's opinion because he did not interview the respondent, the expert's testimony and the prosecutor's comment in closing that the respondent refused an interview did not violate the respondent's right to remain silent. While the facts in Keith are different, Keith supports our view that the use by Kienitz's trial counsel of the lack of an interview to challenge the quality of the expert opinion is inconsistent with the assertion of a Fifth Amendment violation.