Lomholt v. Burt

MEMORANDUM OPINION AND ORDER REGARDING REPORT AND RECOMMENDATION OF MAGISTRATE JUDGE BENNETT, Chief Judge. TABLE OF CONTENTS I.INTRODUCTION. A. Background . B. State Court Proceedings. 1. Pre-trial hearing. 2. The counselor’s testimony. а. Background to the counselor’s testimony. б. Testimony regarding B.G.. c. Testimony regarding N.P.. d. Conclusions regarding both children. 3. The trial court’s ruling. 4. The appellate court’s ruling . 5. Post-conviction relief proceedings. C. The Report and Recommendation. D. Respondent’s Objections. E. Petitioner’s Resistance To Objections. II. LEGAL ANALYSIS. 00 CO A. Standard Of Review For A Report and Recommendation 00 CO B. Standards For Habeas Relief. 05 00 C. Is Lomholt Entitled To Habeas Relief?. O G5 1. “Clearly established Federal law”. O 05 2. “Unreasonable determination” of facts. W Q 3. “Unreasonable application” of law to facts. Ü5 05 *980 4. “Harmless error . Oi G* D. Certificate Of Appealability O o o III. CONCLUSION. .1001 This petition for a writ of habeas corpus comes before the court pursuant to the February 5, 2002, Report and Recommendation by Magistrate Judge Paul A. Zoss on the merits of the petition. Judge Zoss recommends that the petition be granted, that judgment enter in favor of petitioner Lomholt and against the respondent, and that Lomholt be released pending a new trial. The respondent filed objections to the Report and Recommendation on February 14, 2002, and Lomholt filed a resistance to those objections on February 19, 2002. I. INTRODUCTION A. Background Lomholt seeks habeas corpus relief pursuant to 28 U.S.C. § 2254 from his August 16, 1996, conviction by a jury on two counts of second degree sexual abuse, in violation of Iowa Code §§ 709.1 and 709.3. Lomholt was convicted primarily on the testimony of the two victims, B.G., a four-year-old female, and N.P., a five-year-old female, and his own confession. Lomholt contends here, as he did before the Iowa state courts on direct appeal, that he was denied his Sixth Amendment right to confront witnesses at trial when the child victims were permitted to testify outside of his presence via closed-circuit television pursuant to Iowa Code § 910.14(a) (since recodified as § 915.38(1)). B. State Court Proceedings I. Pre-trial hearing Before deciding to allow the child witnesses to testify at trial by closed-circuit television outside of Lomholt’s presence, the trial court held a pre-trial hearing at which the only witness was Patricia A. Tomson, the children’s sex abuse counsel- or. The children did not testify at that hearing. Ms. Tomson’s testimony appears in its entirety in the Iowa Supreme Court Appendix, Sup.Ct. No. 96-1965 (ISC App.), pp. 9-30. Nevertheless, the court will identify the salient parts of her testimony, as her testimony was the basis for the trial court’s conclusions, and hence, is the basis for Lomholt’s contention that the trial court’s ruling violated his Confrontation Clause rights. 2. The counselor’s testimony a. Background to the counselor’s testimony Ms. Tomson described herself as the executive director of the Parents United program, which treats sexually abused children and their families. ISC App. at 9. Ms. Tomson had a master’s degree in mental health counseling, advanced training in sexual abuse treatment and play therapy, and approximately ten years of experience counseling sexually abused children, with a specialty in treating sexually abused children ages three to twelve. Id. at 10. Tomson treated B.G. in group and individual sessions for three to four weeks beginning approximately four months after the alleged incident of abuse by Lomholt, with sessions concluding when B.G.’s family moved away approximately one-and-a-half months before the hearing and approximately three months before Lomholt’s trial. Id. at 25, 19-20. Tomson treated N.P., also in group and individual sessions, beginning about four months after the alleged abuse and continuing through the time of the hearing. Id. at 16, 26. b. Testimony regarding B.G. Tomson testified that various drawings that she had asked B.G. to make of herself, *981 her family, and Lomholt, who was her uncle, showed that B.G. felt “helpless.” Id. at 11-12. Tomson testified that, in the explanation of how she was feeling in one picture, B.G. said “I’m feeling happy because I like my mom, but I feel sad and tired when I think about [Lomholt].” Id. at 12. Tomson also described a session of “sand tray play” in the course of which B.G. identified a “bad character” as Lom-holt, placed him in “jail,” dumped water on him, and cut off his head. Id. at 13-14. Tomson also described a drawing by B.G. of herself missing body parts, made during a group session in which the group discussed the fact that one of the other children needed to go to court. Id. at 14-15. Based upon her education and experience, Tomson described that drawing as showing “that [B.G.] doesn’t have much ego strength and is pretty disintegrated emotionally, particularly when thinking about the subject we were talking about which was court.” Id. at 15. Tomson explained that her sessions with B.G. ended when B.G.’s mother moved away, apparently as the result of her separation from her husband, who was a relative of Lomholt, over what she understood B.G.’s mother considered a lack of support concerning the abuse incident from her husband’s family. Id. at 19-20. The following exchange between the prosecutor and Tomson then occurred: Q I want to discuss [B.G.] right now. Is [B.G.] scared of the defendant? A When I asked her that question directly, she didn’t answer me. She had a relationship with [Lomholt]., He was her care giver. So I’m not certain that she’s frightened of him. * * * * * * Q How about now [B.G.]? Is she apprehensive or scared about testifying in court with [Lomholt] present? A [B.GJs anxious about talking about the abuse at all. I was present when she talked with you about that. And she sat in my lap and did as many distracting things as she could rather than talk about what had happened to her. . The fact that she resorts to baby talk and is very difficult to understand when she starts talking about the abuse. And the fact that she wets her pants or has — now she asks to go [to] the bathroom, but at first, she didn’t. She just wet her pants. Id. at 20-21. On cross-examination by Lomholt’s defense counsel, the following exchange regarding B.G. occurred: Q It’s your testimony that you don’t believe [B.G.] is afraid of [Lomholt]? A She has not indicated that to me. Q And since she broke off any kind of relationship, you really don’t know today what her mental thought pattern is or her ideas are on [Lomholt]? A No, I don’t. Q And would it be fair to say that if I have seen no signs of her being afraid of [Lomholt], it would be less of a burden or less pressure on her to testify in front of [Lomholt]? A It’s very hard for people to testify in front of anyone they have a relationship with. Q Is it possible that they would be— she would be less likely to feel this pressure? A Children — I don’t know if you’re familiar with the accommodation syndrome, but children will make things right. Possibly she would retract anything that she has said because she wouldn’t want to hurt the family or [Lomholt]. ISC App. at 28-29. On redirect examination, the following exchange occurred between the prosecutor and Tomson concerning B.G.: *982 Q You testified that [B.G.] has never told you that she is afraid of [Lomholt]? A That’s right. Q Based upon her drawings and based upon your other communications with her, has she indirectly indicated that she’s afraid of [Lomholt]? A Based on her drawings, there is more indication that [sic] of shame about talking about the abuse than of [Lom-holt] himself. ISC App. at 29. Before noting some testimony by the counselor concerning both children, the court turns to consideration of Tomson’s testimony related specifically to N.P. c. Testimony regarding N.P. Tomson also asked N.P. to draw pictures as part of her treatment process. Tomson testified that N.P. drew two pictures of Lomholt on the same day. In the first of these pictures, N.P. drew Lom-holt’s head and body and what N.P. said was Lomholt’s “peter in the window,” but she testified that N.P. then scribbled out the “peter” explaining, “because he’s naughty.” ISC App. at 16-17. The second drawing apparently took a vei-y long time, because N.P. was restive and resistant to doing it. Id. at 17. When completed, it consisted of Lomholt’s face with a “very big smile,” and N.P. explained that Lomholt was “happy because he touched me all over.” Id. On NJP.’s first visit, Tomson asked her to draw her family, and N.P. complied by drawing herself, but with no facial features, and her brother, but even after being asked if there were other family membei-s, she would not include them, which Tomson concluded indicated a “lack of family support that [N.P.] sees.” Id. at 18. Another drawing of Lomholt that N.P. made showed him “with very big eyes, lots of hair, a mustache and a big smile.” Id. at 19. When asked where his body was, N.P. said “whoops,” then added Lomholt’s body on another piece of paper. Id. As Tomson explained, The bottom part of the body she spent a lot of time drawing this with the eoloi-s. And I asked her what this was and she said penis, no, peter. And then the pants. That seemed to be very significant too for her because she spent so much time on it. Id. at 19. After the prosecutor asked if B.G. was frightened of Lomholt, and received the answer quoted above, the following exchange concerning N.P. occurred: Q How about [N.P.]? A Yes, [N.P.] is frightened of [Lom-holt], Q How do you know this? A She has said that she is. She doesn’t want to see him. Her words are, I want him to stand in a comer for a long, long time. Id. at 20. Similarly, after the prosecutor asked if B.G. was frightened of testifying in Lomholt’s presence, the following exchange concerning N.P. occurred: Q How about [N.P.]? A I was present when [N.P.] talked with you in your office also, and she did not speak up. She talks with me much better in the office than she had talked with you. She, last week, didn’t want to talk about anything about [Lomholt] or draw the picture. She played the baby. She got a baby bottle and drank out of the baby bottle. She does a lot of distracting behaviors also. Id. at 21. On cross-examination, after exploring the effect of “accommodation syndrome” on B.G., the following exchange concerning N.P. occurred between defense counsel and Tomson: *983 Q Now, [N.P.] does not have those same type of — same type of relationship with [Lomholt]; is that correct? A That’s right. Q So on that theory, [N.P.] would have less reason to quote, unquote, lie on the stand? A That’s right. Id. at 29. d. Conclusions regarding both children The rest of Tomson’s testimony concerned her conclusions about both children and inferences to be drawn from a study of the ability of children to testify in front of someone who had done something wrong. As to the two children that Lomholt was charged with abusing, on direct examination by the prosecutor, Tomson offered the following conclusions: Q Do you believe that it would be traumatic for the children to be required to testify in court in front of the defendant? A I believe it would be very traumatic and Pm not sure that either child would talk. Q And on what do you base that feeling? A Their anxiousness. And in talking with me, the first issue in treatment is safety. And when I start dealing with a child, I spend considerable time making a child feel safe both in their environment and in the play therapy room and with me. And that takes time. Q If the children were required to testify in the courtroom in front of the defendant, would they be less likely to tell the truth? A I believe they would be less likely to talk, and I also believe that they would be less likely to tell the truth. Q And, once again, on what do you base that opinion? A I base that opinion on my acquaintance with the two children, but also on some studies that I have read about preschoolers testifying in front of the defendant. Id. at 21-22. The court will return to Tomson’s testimony concerning the studies on which she relied and her conclusions about child witnesses in general below. However, Tomson provided other observations on both children involved in the prosecution of Lomholt, which should be surveyed first. On cross-examination, defense counsel returned to the question of whether the children would be able to testify in front of Lomholt: Q And you said that they [meaning specifically B.G. and N.P.] would have trouble talking. They may have trouble talking in a different room with just myself, [the prosecutor] and the judge present; is that correct? A Yes, they may. Q So there actually may be no difference between that and a courtroom setting? A I believe there would be a difference. Id. at 27. Defense counsel did not ask for any illumination of the basis for Tomson’s conclusion that there would “be a difference,” however. Finally, on redirect examination, the following testimony concerning both girls was elicited: Q Based upon your education, based upon your experience, would testifying in the physical presence of [Lomholt] impair [B.G.’s] and [N.P.’s] ability to • communicate? A Yes, it would. Id. at 29-30. Tomson provided- other, more general testimony. Specifically, Tomson testified about a study in which a supposed defendant broke a glass in front of children of *984 different ages. Id. at 22-23. According to Tomson, the results showed that only 25 percent of the children under the age of nine would testify to the fact that the glass was broken by the defendant in the defendant’s presence, but 69 percent of the children over age nine were able to identify the offense in the presence of the offender. Id. at 23. Upon refreshing her memory based on a letter she had written to the prosecutor concerning the study, Tomson testified that 69 percent of the children under age nine would disclose the transgression in the absence of the defendant, but only 25 percent would disclose it in the defendant’s presence. Id. at 24. At the conclusion of Tomson’s direct examination, the prosecutor elicited the following general opinions from Tomson: Q Now, I’m going to ask you an expert opinion not dealing with these two children specifically, but with children of the ages of four and five. In your expert opinion, based upon your education and experience, would it be traumatic for a four or a five year old to be required to testify in front of someone who has sexually abused them? A Yes, in my expert opinion, it would be traumatic for them to testify in his presence. Q For all children? A For all children. Q Can you think of any time when there would ever be an exception to that? A I can’t. Id. at 24-25. Defense counsel took Tom-son up on this opinion on cross-examination, as follows: Q You just stated under direct examination that you know of no scenario for a four or five year old should basically have direct testimony in a courtroom? A Yes. Um, I’ve been involved with 19 children and in some type of court proceedings, several of them were under age six. And it has been extremely traumatic for those children even to think about testifying. Id. at 26. Defense counsel then elicited testimony from Tomson to the effect that she agreed that it was possible that both children involved in this case had gotten the notion that Lomholt was a “bad person,” for example, from their contact with persons investigating the alleged abuse. Id. at 26-27. Turning to the study upon which Tomson had in part relied, Tomson indicated in response to a question from defense counsel that, without the whole study, she could not provide a breakdown of the responses of the 31 percent of children under age nine who did not identify the transgressor in his absence. Id. at 28. 3. The trial court’s ruling In its ruling on the prosecution’s motion to permit testimony of the two child witnesses by closed-circuit television, the trial court found the following facts on the basis of the pre-trial hearing: In support of its motion, the State offered the testimony of Patricia Tom-son, executive director of Parents United of North Central Iowa. Ms. Tomson has a master’s degree in counseling, with a specialty in child victims of sexual abuse. She has provided therapy in this area for more than ten years. Ms. Tomson testified that B.G. was brought in by her mother for treatment during the last week of April, 1996. Tomson saw B.G. several times over the next three or four weeks, individually and in group sessions. She last saw B.G. on May 19, 1996, after which time B.G. and her mother apparently moved from the area. According to Tomson, B.G. becomes very anxious when talking about the incident and initially wet her pants, which Tomson indicated was common with vie- *985 tims. According to Tomson, B.G. told her that she felt “sad and tired” when she thought about the defendant. Tom-son also opined that B.G.’s drawings indicate a feeling of powerlessness. B.G. will sometimes resort to baby talk, rather than discuss the incident. Tomson first met with N.P. on May 9, 1996, and continues to see her on a regular basis. According to Tomson, N.P. is frightened of the defendant and “plays like a baby” when an effort is made to discuss the incident. According to Tomson, it would be “very traumatic” for either of the children to testify in the defendant’s presence. Tomson expressed doubt as to whether either child would talk about the incident and opined, based upon pri- or studies, that it was less likely that they would tell the truth about the incident with the defendant present. When asked directly by the prosecuting attorney, Ms. Tomson testified that the trauma caused by testifying in the physical presence of the defendant would impair the children’s ability to testify. ISC App. at 4-5; see also Appendix of Decisions, Exhibit A, at 4 — 5. After citing the Iowa statute authorizing testimony of child abuse victims by closed-circuit television, Iowa Code § 910A.14(1) (since recodi-fied at § 915.38(1)), and the United States Supreme Court’s standards for safeguarding a defendant’s Confrontation Clause rights before permitting testimony in that way, as set out in Maryland v. Craig, 497 U.S. 836, 110 S.Ct. 3157, 111 L.Ed.2d 666 (1990), the trial court concluded as follows: In the instant action, the State produced credible testimony that testifying in the physical presence of the defendant would be traumatic to each of the alleged victims. In addition, the evidence was convincing that the trauma experienced in testifying would impair the ability of the witnesses to communicate. The court finds that testimony by closed circuit equipment is necessary to protect the alleged victims from trauma. Id. at 6; see also Appendix of Decisions, Exhibit A, at 6. At trial, the children testified by closed-circuit television and the jury also heard evidence concerning Lomholt’s confession. On August 16, 1996, the jury convicted Lomholt of abusing both children. The state trial court denied Lomholt’s motion for a new trial and directed that judgment be entered against him on November 5, 1996. 4. The appellate court’s ruling On January 28, 1998, on Lomholt’s direct appeal of his conviction, which was based primarily on his contention that his rights under the Confrontation Clause had been violated by allowing the child witnesses to testify outside of his presence, the Iowa Court of Appeals affirmed the trial court’s decision to allow the children to testify by closed-circuit television and, consequently, affirmed Lomholt’s conviction. After citing the applicable standards from Maryland v. Craig, 497 U.S. 836, 110 S.Ct. 3157, 111 L.Ed.2d 666 (1990), and Coy v. Iowa, 487 U.S. 1012, 108 S.Ct. 2798, 101 L.Ed.2d 857 (1988), including the fact that the exception to the confrontation right in a case such as this must be based upon case-specific findings, the Iowa Court of Appeals provided the following summary of the evidence and its conclusions: Patricia Tomson, the executive director of Parents United of North Central Iowa, had engaged in group and individual counseling with both children. Tom-son testified B.G. was brought in for treatment by her mother during the last week of April 1996. Tomson visited with B.G. several times during the following three or four weeks, both individually, and in group sessions. Tomson testified B.G. became anxious when de *986 scribing the abuse and wet her pants on one occasion. She stated B.G. indicated she felt “sad and tired” when she thought of Lomholt. Her drawing of herself did not include arms which Tom-son stated was an indication of powerlessness. Tomson initially met with N.P. in May 1996 and continues to see her regularly. Tomson maintained N.P. was afraid of Lomholt and played like a baby when the subject was broached. Tomson testified N.P. said she was frightened of Lomholt and that “she doesn’t want to see him. Her words are, I want him to stand in a corner for a long, long time.” Tomson further testified it would be “very traumatic” for either child to testify' in Lomholt’s presence. She stated, “I’m not sure either child would talk.” Tomson also thought testifying before Lomholt would impair both girls’ ability to testify and make it less likely they would tell the truth. Lomholt maintains Tomson’s testimony that she knew of no situation in which a four or five-year-old child should give direct testimony in the courtroom revealed a predisposed bias specifically applied to N.P. and B.G. While we recognize Tomson did make some generalized statements, she also gave particularized assessments as to each girl based upon numerous counseling sessions with the girls. We find there was specific information presented to the satisfaction of the court to indicate the girls’ ability to testify would be impaired if not shielded from Lomholt. Appendix of Decisions, Exhibit B at 4-5. Additionally, the Iowa Court of Appeals reached the following conclusion: Independent of the girls’ testimony, Lomholt admitted he had been abusing his niece, B.G., “[f]or the last couple months.” Two officers testified about Lomholt’s admissions of the abuse during interviews with him. Lomholt’s written statement indicating he “asked [B.G.] and [N.P.] to pull their pants down and I touched them with my hand and penis and they put it in their mouth[s]” was introduced against him at trial. Lomholt had the opportunity to commit the abuse as he was alone with the girls. In addition, N.P.’s mother testified that N.P.’s personality had changed after spending the evening with Lomholt. These facts collectively corroborate Lomholt’s criminal acts. Appendix of Documents, Exhibit B at 5. The Iowa Supreme Court denied further review on April 3, 1998, and procedendo issued on April 20,1998. 5. Post-conviction relief proceedings The state district court also rejected Lomholt’s application for post-conviction relief, which he filed January 25, 1999. It does not appear that the Iowa district court considered any Confrontation Clause issue in its ruling dismissing Lomholt’s application for post-conviction relief; rather, the court considered various claims of ineffective assistance of counsel, none of which pertained directly or indirectly to the admission of the testimony of the child witnesses via closed-circuit television. See Appendix of Decisions, Exhibit C. The Iowa Supreme Court dismissed Lomholt’s post-conviction relief appeal as frivolous on September 28, 2000. Procedendo on Lom-holt’s post-conviction relief application issued on October 12, 2000. At that point, Lomholt had exhausted his state remedies. C. The Report and Recommendation Lomholt filed the present petition for federal habeas corpus relief on March 16, 2001, asserting several grounds for relief. However, after consulting with counsel, Lomholt eventually withdrew all grounds for relief except his contention that the *987 trial court violated his Sixth Amendment right to confront witnesses by allowing the two minor children to testify at his trial via closed-circuit television. The matter proceeded to briefing on the merits of that issue, and on February 5, 2002, Magistrate Judge Paul A. Zoss filed a Report and Recommendation on its disposition. In his Report and Recommendation, Judge Zoss recommends that Lomholt’s petition for federal habeas corpus relief be granted. Although Judge Zoss noted that the state trial court had based its decision to allow the two children to testify via closed-circuit television on evidence taken at a pre-trial hearing, Judge Zoss concluded that the trial court’s decision to permit the children to testify in this way constituted both an unreasonable determination of the facts and an unreasonable application of United States Supreme Court precedent to those facts. More specifically, Judge Zoss found that the record before the trial court revealed that B.G., the younger victim and Lom-holt’s niece, felt “sad” when she thought about Lomholt, but was not frightened of him. Although the record before the trial court revealed that N.P., the other victim, who was not related to Lomholt, was frightened of Lomholt, Judge Zoss concluded that there was no specific evidence indicating that she would have more difficulty testifying about Lomholt’s conduct in his presence than in his absence. Although the counselor testified that it would be “very traumatic” for both children to testify in Lomholt’s presence, because the children were “anxious,” Judge Zoss concluded that there is nothing in the record to demonstrate that their “anxiousness” resulted from the prospect of having to testify in Lomholt’s presence, as opposed to having to testify in a courtroom, or that any trauma from testifying in Lomholt’s presence would be more than de minimis. Therefore, Judge Zoss concluded that none of the evidence presented at the pre-trial hearing was sufficient to support the trial court’s finding that either child would be traumatized any more by testifying in Lomholt’s physical presence than she would be by testifying in a courtroom generally, or that any such additional trauma would be significant rather than de min-imis. Judge Zoss also concluded that the counselor’s testimony that all four- and five-year-old children would be traumatized by testifying in front of someone who had sexually abused them “seems to be the real crux of [the counselor’s] testimony,” but that, even if the counselor’s conclusion was justified, it was not the sort of “case-specific” finding required by Maryland v. Craig, 497 U.S. 836, 110 S.Ct. 3157, 111 L.Ed.2d 666 (1990), the Supreme Court’s governing precedent on the question. Report and Recommendation at 20. Judge Zoss concluded that this was so, not least, because he found that the counselor “testified that one of the child witnesses, B.G., was not afraid of the defendant.” Id. (emphasis in the original). Judge Zoss concluded, further, that the Confrontation Clause violation could not be disregarded as “harmless error.” Judge Zoss reasoned that, if the children’s testimony were excluded from the record, the conviction could not be based on Lomholt’s confession, because, even if his confession was totally uncoerced, there was no corroborating evidence that the crime occurred. D. Respondent’s Objections In his Objections, filed February 14, 2002, the respondent objects, first, to Judge Zoss’s conclusion that the Iowa courts made an unreasonable determination of the facts. The respondent contends that the evidence presented by the counselor at the pre-trial hearing is sufficient *988 to support findings- that both B.G. and N.P. would suffer sufficient trauma from testifying in court and in front of Lomholt to impair their ability to communicate. The respondent points out that, when asked specifically whether testifying in front of Lomholt would impair the children’s ability to communicate, the counsel- or answered in .the affirmative as to both children, based on her education, training, and case-specific observations of the children. The respondent argues that, even if the counselor’s case-specific conclusions or her categorical conclusion that all children of the age of the victims in this case would be traumatized by testifying in the presence of their alleged abuser might be open to challenge by a contrary expert opinion, Lomholt did not present any such evidence to the trial court to undermine the basis for the trial court’s decision. Thus, the respondent contends that the trial court’s determination of the facts was not unreasonable in light of the evidence presented. Second, the respondent objects to Judge Zoss’s conclusion that the Iowa courts unreasonably applied Maryland v. Craig to the facts in Lomholt’s case. The respondent argues that, if this court rejects Judge Zoss’s conclusion that the trial court’s findings were unreasonable, it must necessarily conclude that the application of the law to the facts was not unreasonable, either, because the findings satisfy the constitutional requirements. The respondent contends, further, that Judge Zoss’s reliance on Hoversten v. Iowa, 998 F.2d 614 (8th Cir.1993), in which the court held that the constitutional standard had not been satisfied, is misplaced, because in that case, the trial court permitted use of a one-way mirror without any pre-trial hearing whatsoever. Nor, the respondent argues, was it improper, under other circuit precedent cited by Judge Zoss, for the trial court to reach its conclusion to permit testimony by closed-circuit television without hearing pre-trial testimony of the children in chambers. Finally, the respondent objects to Judge Zoss’s conclusion that any error was not “harmless.” In the respondent’s view, Lomholt’s confession was sufficiently corroborated by other evidence, including N.P.’s mother’s testimony about N.P.’s behavior change after the alleged abuse, to sustain the conviction, even in the absence of testimony by the children. Furthermore, the respondent argues, N.P.’s trial testimony was untainted by any Confrontation Clause violation, even if there was a Confrontation Clause violation as to B.G.’s trial testimony, and thus, N.P.’s untainted testimony was sufficient to corroborate Lomholt’s confession. E. Petitioner’s Resistance To Objections In his Resistance to Objections to Report and Recommendation, Lomholt contends that there was no credible evidence that Lomholt’s presence would traumatize either girl or prevent either girl from being able to testify. Moreover, Lomholt contends that any Confrontation Clause violation was not “harmless,” because the supposedly corroborating evidence of N.P.’s behavior change after the alleged abuse was inadmissible hearsay, and N.P.’s trial testimony also constituted a Confrontation Clause violation, such that it cannot be used to corroborate Lomholt’s confession. II. LEGAL ANALYSIS A. Standard Of Review For A Report and Recommendation The standard of review to be applied by the district court to a report and recommendation of a magistrate judge is established by statute: *989 A judge of the court shall make a de novo determination of those portions of the report or specified proposed findings or recommendations to which objection is made. A judge of the court may accept, reject, or modify, in whole or in part, the findings or recommendations made by the magistrate [judge]. 28 U.S.C. § 686(b)(1). The Eighth Circuit Court of Appeals has repeatedly held that it is reversible error for the district court to fail to conduct a de novo review of a magistrate judge’s report where such review is required. See, e.g., Hosna v. Groose, 80 F.3d 298, 306 (8th Cir.) (citing 28 U.S.C. § 636(b)(1)), cert. denied, 519 U.S. 860, 117 S.Ct. 164, 136 L.Ed.2d 107 (1996); Grinder v. Gammon, 73 F.3d 793, 795 (8th Cir.1996) (citing Belk v. Purkett, 15 F.3d 803, 815 (8th Cir.1994)); Hudson v. Gammon, 46 F.3d 785, 786 (8th Cir.1995) (also citing Belk). However, the plain language of the statute governing review provides only for de novo review of “those portions of the report or specified proposed findings or recommendations to which objection is made.” 28 U.S.C. § 636(b)(1). Therefore, portions of the proposed findings or recommendations to which no objections are filed are reviewed only for “plain error.” See Griffini v. Mitchell, 31 F.3d 690, 692 (8th Cir.1994) (reviewing factual findings for “plain error” where no objections to the magistrate judge’s report were filed). The court finds that the respondent’s objections here are sufficiently specific to invoke de novo review of the challenged portions of the Report and Recommendation. Such de novo review in this case necessarily begins with the standards generally applicable to habeas corpus relief from a state conviction. B. Standards For Habeas Relief As the Eighth Circuit Court of Appeals recently explained, The Antiterrorism and Effective Death Penalty Act (AEDPA) mandates that habeas relief “shall not be granted with respect to any claim that was adjudicated on the merits in State court proceedings unless” the state court’s decision was “contrary to, or involved an unreasonable application of, clearly established Federal law, as determined by the Supreme Court of the United States,” or “was based on an unreasonable determination of the facts in light of the evidence presented in the State court proceeding.” 28 U.S.C. § 2254(d). Robinson v. Crist, 278 F.3d 862, 865 (8th Cir.2002); accord Sexton v. Kemna, 278 F.3d 808, 811 (8th Cir.2002). Section 2254(d) thus “ ‘mandates a deferential review of state court decisions.’ ” Boyd v. Minnesota, 274 F.3d 497, 500 (8th Cir.2001) (quoting James v. Bowersox, 187 F.3d 866, 869 (8th Cir.1999)). More specifically, as to the alternatives under § 2254(d)(1), The Supreme Court recently clarified the proper interpretation [of] § 2254(d)(1). Writing for a majority of the Court on this issue, Justice O’Con-nor explained that the “contrary to” clause limits habeas relief to cases in which “the state court arrives at a conclusion opposite to that reached by this Court on a question of law or ... decides a case differently than this Court has on a set of materially indistinguishable facts.” The “unreasonable application” clause permits the writ to issue only “if the state court identifies the correct governing legal principle from this Court’s decisions but unreasonably applies that principle to the facts of the prisoner’s case.” Williams v. Taylor, 529 U.S. 362, 402-13, 120 S.Ct. 1495, 146 L.Ed.2d 389 (2000). Sexton, 278 F.3d at 811. As to the “unreasonable determination” alternative under *990 § 2254(d)(2), the Eighth Circuit Court of Appeals has explained that “a state court’s determination on the merits of a factual issue is entitled to a presumption of correctness.” Boyd, 274 F.3d at 500 (citing 28 U.S.C. § 2254(e)(1)). The petitioner has the “burden of rebutting the presumption of correctness by clear and convincing evidence.” 28 U.S.C. § 2254(e)(1); see also Kinder v. Bowersox, 272 F.3d 532, 541 (8th Cir.2001). Thus, if “there is sufficient record evidence to support such a finding ..., it would not constitute an unreasonable determination of the facts in light of the evidence presented at trial,” Boyd, 274 F.3d at 501 n. 4, or, presumably, at a pretrial evidentiary hearing, which is what is at issue here. In short, “ ‘mere disagreement with the [state] court’s conclusions is not enough to warrant habeas relief.’ ” Boyd, 274 F.3d at 500 (quoting Long v. Humphrey, 184 F.3d 758, 761 (8th Cir.1999), in turn quoting Matteo v. Superintendent SCI Albion, 171 F.3d 877, 890 (3rd Cir.1999)). Rather, the state court’s decision must not only be “wrong,” it must be “unreasonable.” See, e.g., Williams v. Taylor, 529 U.S. 362, 410, 120 S.Ct. 1495, 146 L.Ed.2d 389 (2000) (“[T]he most important point is that an unreasonable application of federal law is different from an incorrect application of federal law.”) (emphasis in the original); Kinder, 272 F.3d at 541 (“As with ‘reasonableness’ in evaluating the application of clearly established law, that a federal ha-beas court might believe the findings of the state court to be incorrect does not mean they are unreasonable under § 2254(d)(2).”) (citing Weaver v. Bowersox, 241 F.3d 1024, 1030 (8th Cir.2001)). C. Is Lomholt Entitled To Habeas Relief? In light of the general standards for federal habeas corpus relief from a state conviction and the respondent’s objections, it is apparent that the questions upon de novo review of the challenged portions of Judge Zoss’s Report and Recommendation in this case include the following: (1) What is the “clearly established Federal law” against which the trial court’s determinations must be measured? (2) Was the trial court’s factual determination unreasonable? (3) Was the trial court’s application of “clearly established Federal law” to those facts unreasonable? and (4) Was any Confrontation Clause violation nevertheless only “harmless error”? The court will consider these question in turn. 1. “Clearly established Federal law” Judge Zoss found, and the parties do not disagree, that the “clearly established Federal law, as determined by the Supreme Court of the United States,” see 28 U.S.C. § 2254(d)(1), on the Confrontation Clause question presented here is embodied in the Supreme Court’s decision in Maryland v. Craig, 497 U.S. 836, 110 S.Ct. 3157, 111 L.Ed.2d 666 (1990). This court also agrees that Craig sets the legal standards in this case. In Craig, the Court was called upon “to decide whether the Confrontation Clause of the Sixth Amendment categorically prohibits a child witness in a child abuse case from testifying against a defendant at trial, outside of the defendant’s physical presence, by one-way closed circuit television.” Craig, 497 U.S. at 840, 110 S.Ct. 3157. Justice O’Connor, writing for a five-four majority, concluded that it did not. The Court in Craig concluded, after extensive analysis, that a face-to-face confrontation requirement, while forming “ ‘the core of the values furthered by the Confrontation Clause,’ ” see Craig, 497 U.S. at 847, 110 S.Ct. 3157 (quoting California v. Green, 399 U.S. 149, 157, 90 S.Ct. 1930, 26 L.Ed.2d 489 (1970)), nevertheless “is not absolute.” Id. at 850, 110 S.Ct. 3157. On the other hand, the fact that the *991 right is not absolute “does not, of course, mean that it may easily be dispensed with.” Id. at 850, 110 S.Ct. 3157. Rather, the Court concluded that “our precedents confirm that a defendant’s right to confront accusatory witnesses may be satisfied absent a physical, face-to-face confrontation at trial only where denial of such confrontation is necessary to further an important public policy and only where the reliability of the testimony is otherwise assured.” Id. (emphasis added). The Court then held that, “if the State makes an adequate showing of necessity, the state interest in protecting child witnesses from the trauma of testifying in a child abuse case is sufficiently important to justify the use of a special procedure that permits a child witness in such cases to testify at trial against a defendant in the absence of face-to-face confrontation with the defendant.” Id. at 855, 110 S.Ct. 3157. Turning to the “requisite finding of necessity,” the Court explained in Craig that the finding “must of course be a case-specific one: The trial court must hear evidence and determine whether use of the one-way closed circuit television procedure [authorized by a Maryland statute at issue in that case] is necessary to protect the welfare of the particular child witness who seeks to testify.” Id. Furthermore, “[t]he trial court must also find that the child witness would be traumatized, not by the courtroom generally, but by the presence of the defendant.” Id. at 856, 110 S.Ct. 3157. Emphasizing this second point further, the Court explained, Denial of face-to-face confrontation is not needed to further the state interest in protecting the child witness from trauma unless it is the presence of the defendant that causes the trauma. In other words, if the state interest were merely the interest in protecting child witnesses from courtroom trauma generally, denial of face-to-face confrontation would be unnecessary because the child could be permitted to testify in less intimidating surroundings, albeit with the defendant present. Finally, the trial court must find that the emotional distress suffered by the child witness in the presence of the defendant is more than de minimis, i.e., more than “mere nervousness or excitement or some reluctance to testify,” [citations omitted]. We need not decide the minimum showing of emotional trauma required for use of the special procedure, however, because the Maryland statute, which requires a determination that the child witness will suffer “serious emotional distress such that the child cannot reasonably communicate,” [citation omitted], clearly suffices to meet constitutional standards. Craig, 497 U.S. at 856, 110 S.Ct. 3157. The Court summarized its conclusions on the circumstances in which an exception to face-to-face confrontation is permissible, as follows: [W]here necessary to protect a child witness from trauma that would be caused by testifying in the physical presence of the defendant, at least where such trauma would impair the child’s ability to communicate, the Confrontation Clause does not prohibit use of a procedure that, despite the absence of face-to-face confrontation, ensures the reliability of the evidence by subjecting it to rigorous adversarial testing and thereby preserves the essence of effective confrontation. Because there is no dispute that the child witnesses in this case testified under oath, were subject to full cross-examination, and were able to be observed by the judge, jury, and defendant as they testified, we conclude that, to the extent that a proper finding of necessity has been made, the admis *992 sion of such testimony would be consonant with the Confrontation Clause. Craig, 497 U.S. at 857, 110 S.Ct. 3157. The Court in Craig then turned to the question of whether a proper finding of necessity had been made in the case before it. The Court observed that the Maryland Court of Appeals had concluded that, to invoke the procedure for testimony of a child sex abuse victim via one-way closed-circuit television, as permitted by a Maryland statute — and as permitted by the Iowa statute now at issue in this case— “the Confrontation Clause required the trial court to make a specific finding that testimony by the child in the courtroom in the presence of the defendant would result in the child suffering serious emotional distress such that the child could not reasonably communicate.” Id. at 858, 110 S.Ct. 3157 (citing the decision of the Maryland court) (emphasis in the original). “This conclusion,” the Supreme Court concluded, “of course, is consistent with our holding today.” Id. However, the Court rejected the Maryland court’s holdings that the child witness must initially be questioned in the defendant’s presence and that the trial judge must determine whether a child would suffer severe emotional distress if he or she were to testify by two-way closed-circuit television, before permitting use of the one-way television procedure authorized by the statute. Id. The Court explained, Although we think such evidentiary requirements could strengthen the grounds for use of protective measures, we decline to establish, as a matter of federal constitutional law, any such categorical evidentiary prerequisites for the use of the one-way television procedure. The trial court in this case, for example, could well have found, on the basis of the expert testimony before it, that testimony by the child witnesses in the courtroom in the defendant’s presence “will result in [each] child suffering serious emotional distress such that the child cannot reasonably communicate,” [citations omitted]. So long as a trial court makes such a case-specific finding of necessity, the Confrontation Clause does not prohibit a State from using a one-way closed circuit television procedure for the receipt of testimony by a child witness in a child abuse case. Craig, 497 U.S. at 860, 110 S.Ct. 3157. Because the Supreme Court could not be sure that the Maryland court would have reached the same conclusion — that the trial court had not made the requisite finding of necessity — had the court applied the legal standard that the Supreme Court had established in its ruling, the Supreme Court vacated the judgment of the Maryland Court of Appeals and remanded the case for further proceedings. Id. 2. “Unreasonable determination” of facts Had Judge Zoss been the original finder of fact on the question of the extent of the “trauma” the child witnesses against Lomholt would suffer from testifying in Lomholt’s presence, 'this court would undoubtedly have concluded that Judge Zoss’s findings were reasonable and this court would not have hesitated to uphold them. In short, this court agrees with Judge Zoss that the trial court’s findings were “wrong,” or at least, were based on evidence that this court would not find satisfactory if this court were the finder of fact. However, in light of the standards set out above, on a petition for habeas corpus relief from a state conviction, this court cannot simply disagree with the state court, nor upon de novo review of a magistrate judge’s report and recommendation in such a context can the court simply embrace the magistrate judge’s conclusion from the record established in state court. Rather, this court must defer to the state *993 court’s determination of the facts, and can only grant relief if the state court’s determination of the facts was not only “wrong,” but “unreasonable.” See 28 U.S.C. § 2254(d)(2) & (e)(1); Boyd, 274 F.3d at 500; Kinder, 272 F.3d at 541. Here, based on a review of the record, and in light of the presumption of correctness to which the trial court’s factual determinations are entitled, see Boyd, 274 F.3d at 500 (citing 28 U.S.C. § 2254(e)(1)), the court cannot say that the trial court’s findings are so lacking in evidentiary support as to be “unreasonable.” See Boyd, 274 F.3d at 501 n. 4. The “requisite showing of necessity” to tolerate an exception to the Confrontation Clause “required the trial court to make a specific finding that testimony by the child in the courtroom in the presence of the defendant would result in the child suffering serious emotional distress such that the child could not reasonably communicate.” Craig, 497 U.S. at 858, 110 S.Ct. 3157 (emphasis in the original); see also id. at 855-56, 110 S.Ct. 3157 (the procedure is necessary only if a case-specific finding is made that the child witness will suffer from trauma caused by testifying in the physical presence of the defendant such that, at a minimum, the trauma would impair the child’s ability to communicate). Indeed, the Iowa statute permitting testimony of child abuse victims by closed-circuit television mirrors these factual predicates, because it also requires that the trial court find that a minor must be protected “from trauma caused by testifying in the physical presence of the defendant where it would impair the minor’s ability to communicate,” and “such an order shall be entered only upon a specific finding by the court that such measures are necessary to protect the minor from trauma.” Iowa Code § 910.14(1) (now Iowa Code § giS.SSU)). 1 This court must presume the correctness of the state court’s determination on the merits of the factual issues required by the statute and the Confrontation Clause, see Boyd, 274 F.3d at 500 (citing 28 U.S.C. § 2254(e)(1)); the petitioner has the “burden of rebutting the presumption of correctness by clear and convincing evidence,” see 28 U.S.C. § 2254(e)(1); see also Kinder, 272 F.3d at 541; and this court must conclude that the *994 trial court’s factual determination was not “unreasonable” if “there is sufficient record evidence to support such a finding.” Boyd, 274 F.3d at 501 n. 4. There is plainly sufficient evidence in the record to support the trial court’s conclusion that the children would suffer trauma from testifying. The counselor’s unre-butted testimony clearly supports such a conclusion. For example, as noted above, the counselor testified, specifically as to each child, that the children would avoid the topic of the abuse and engage in distracting, “baby” behavior rather than discuss it; that both children’s behavior indicated that they felt “powerless”; and that B.G.’s drawing of herself missing body parts made in the course of a discussion about another child in the group session needing to go to court demonstrated “that she doesn’t have much ego strength and is pretty disintegrated emotionally, particularly when thinking about the subject we were talking about which was court.” ISC App. at 15. Also, in response to defense counsel’s question about whether there would be a difference for the children between talking in a different room with just defense counsel, the prosecutor, and the judge present, and talking in “a courtroom setting,” the counselor opined that she “believe[d] there would be a difference.” Id. at 27. What is admittedly more uncertain is the sufficiency of the evidence to establish that the children would suffer trauma from testifying in Lomholt’s presence, not just from testifying in court. However, as the respondent points out, when the counselor was asked specifically, “Do you believe that it would be traumatic for the children to be required to testify in court in front of the defendant?” her answer was not just “yes,” but that it would be “very traumatic,” id. at 21 (emphasis added), i.e., not merely de minimis. She explained, further, that this conclusion was based on “[tjheir anxiousness.” Id. Moreover, she testified that N.P. was “frightened of [Lomholt],” which she testified she knew from N.P.’s express statement to that effect, and that she did not want to see him, including her words that “I want him to stand in a corner for a long, long time.” Id. at 20. On the other hand, Tomson also responded to the prosecutor’s reiteration of the question of whether B.G. was “apprehensive or scared about testifying in court with [Lomholt] present” by stating that “[B.G.]’s anxious about talking about the abuse at all.” Id. at 20-21. Later in the hearing, when asked if B.G. had “indirectly indicated that she’s afraid of [Lomholt],” Tomson testified that, “[b]ased on her drawings, there is more indication that [sic] of shame about talking about the abuse than of [Lomholt] himself.” Id. at 29. Neither of these responses suggests that Lo'mholt’s presence was the source of greater trauma than testifying in court generally. Nevertheless, while the record is somewhat equivocal on the question of whether B.G. was “afraid” of Lomholt, the court cannot embrace Judge Zoss’s characterization of the record regarding B.G. as including specific testimony by Tomson that B.G. “was not afraid of the defendant.” Report and Recommendation at 20 (emphasis in the original). The portions of Tomson’s testimony that Judge Zoss cites in support of this conclusion show, instead, that when Tomson asked B.G. if she was scared of the defendant, “she didn’t answer me,” and that, because B.G. had a relationship with Lomholt as her caregiver, Tomson was “not certain that she’s frightened of him,” ISC App. at 20 (emphasis added), and that B.G. “has not indicated that to me.” Id. at 28. These statements are not the same as statements that B.G. was not afraid of Lomholt. *995 Moreover, although the court agrees that neither B.G. nor N.P. displayed the degree of “fear” that sustained the federal trial court’s decision to allow the child witnesses to testify by closed-circuit television in United States v. Rouse, 111 F.3d 561, 568-69 (8th Cir.), cert. denied, 522 U.S. 905, 118 S.Ct. 261, 139 L.Ed.2d 188 (1997), or the state trial court’s similar determination in LaBayre v. Iowa, 97 F.3d 1061, 1062-63 (8th Cir.1996), cert. denied, 519 U.S. 1136, 117 S.Ct. 1003, 136 L.Ed.2d 882 (1997), the court finds nothing in Craig suggesting that the “trauma” from testifying in the defendant’s presence must necessarily equate with “fear” of the defendant. Rather, the Court in Craig framed the standard in terms of “trauma” and “emotional distress.” See Craig, 497 U.S. at 857, 110 S.Ct. 3157 (“trauma”) & 860 (approving the Maryland statutory formulation in terms of “serious emotional distress”). There is significant evidence in the record that B.G. would be “emotionally distressed” or “traumatized” by testifying in Lomholt’s presence, even if she was not “afraid” of him, based on Tomson’s testimony that B.G. avoided answering the question of whether she was afraid of Lomholt, and Tomson’s testimony concerning the pressure that B.G. would feel in testifying against someone with whom she had a relationship, like Lomholt, and the effects of “accommodation syndrome” in the circumstances of this case, which Tom-son testified might cause B.G. to “retract anything that she has said because she wouldn’t want to hurt the family or [Lom-holt].” Id. at 28-29. This evidence does suggest “trauma” to B.G. from testifying in Lomholt’s presence, even if B.G. was not actually “afraid” of Lomholt. Similarly, the fact that Tomson conceded that B.G. and N.P. may have obtained an impression that Lomholt is a “bad person” from other persons who had investigated the alleged abuse does not necessarily establish that their “trauma” or “emotional distress” at testifying in front of Lomholt was not real or should be disregarded. Thus, while this court might not have made such a finding on the record presented, in light of the deference and presumption of correctness to which the trial court’s factual determination of “trauma” is entitled, see Boyd, 274 F.3d at 500 (citing 28 U.S.C. § 2254(e)(1)), the court cannot say that the trial court’s finding that there was “credible testimony that testifying in the physical presence of the defendant would be traumatic to each of the alleged victims,” ISC App. at 6; see also Appendix of Decisions, Exhibit A, at 6, was so lacking in evidentiary support in the record that it was “unreasonable.” See Boyd, 274 F.3d at 501 n. 4. To put it another way, the court cannot find that Lomholt has rebutted the presumption of correctness of the trial court’s finding of “trauma” with the necessary “clear and convincing evidence.” See 28 U.S.C. § 2254(e)(1); Kinder, 272 F.3d at 541. Yet “trauma” alone is not enough. Under both the constitutional and statutory standards, the trial court was also required to find that the trauma was not merely de minimis, but would be such that it would impair the children’s ability to communicate. See Craig, 497 U.S. at 857, 110 S.Ct. 3157 (“Finally, the trial court must find that the emotional distress suffered by the child witness in the presence of the defendant is more than de minimis, i.e., more than ‘mere nervousness or excitement or some reluctance to testify,’ [citations omitted]. We need not decide the minimum showing of emotional trauma required for use of the special procedure, however, because the Maryland statute, which requires a determination that the child witness will suffer ‘serious emotional distress such that the child cannot reasonably com *996 municate,’ [citation omitted], clearly suffices to meet constitutional standards.”); 2 Iowa Code § 910A.14(1) (the trial court must find that a minor must be protected “from trauma caused by testifying in the physical presence of the defendant where it would impair the minor’s ability to communicate”). On this point, the record is also sufficient to sustain the trial court’s finding that “the evidence was convincing that the trauma experienced in testifying would impair the ability of the witnesses to communicate,” ISC App. at 6; see also Appendix of Decisions, Exhibit A, at 6, at least under the standards to which this court must adhere on habeas corpus review. Tomson testified, in response to questions concerning the effect of Lomholt’s presence on the children, that she did not think either child would talk if required to testify in his presence, or if the children did talk, that they would tell the truth. See ISC App. at 21-22 & 29-80. Again, Lomholt has failed to rebut the presumption of correctness of the trial court’s finding that the trauma from testifying in Lomholt’s presence would be such that it would impair the children’s inability to communicate with the necessary “clear and convincing evidence.” See 28 U.S.C. § 2254(e)(1); Kinder, 272 F.3d at 541. This court must, therefore, sustain the respondent’s first objection to the February 5, 2002, Report and Recommendation, and reject Judge Zoss’s conclusion that the trial court made an “unreasonable determination” of the facts in light of the record presented. 3. “Unreasonable application” of law to facts The respondent also objects to Judge Zoss’s conclusion that the trial court made an “unreasonable application” of federal law, as embodied in United States Supreme Court precedent, to the facts in the record. See 28 U.S.C. § 2254(d)(1). There does not appear to be any contention that the trial court “incorrectly identified the governing legal principle from [the United States Supreme] Court’s decisions.” See Williams, 529 U.S. at 413, 120 S.Ct. 1495; Sexton, 278 F.3d at 811. Rather, as noted above, the trial court expressly—and correctly—relied on Maryland v. Craig, 497 U.S. 836, 110 S.Ct. 3157, 111 L.Ed.2d 666 (1990), as stating the applicable legal standard. Thus the question is whether the trial court “unreasonably applie[d] [the Craig ] principiéis] to the facts of [Lomholt’s] case.” Williams, 529 U.S. at 413, 120 S.Ct. 1495. Again, the state court’s determination must not only be “wrong,” it must be “unreasonable.” See Williams, 529 U.S. at 410, 120 S.Ct. 1495 (“[T]he most important point is that an unreasonable application of federal law is different from an incorrect application of federal law.”) (emphasis in the original). This court must again respectfully disagree with Judge Zoss’s conclusion that the trial court’s application of federal law was “unreasonable,” and not merely “wrong,” although this court might agree with Judge Zoss that the trial court’s determination was the latter. First, although no party argues otherwise, it is worth noting that there is no apparent difference between the constitutional standard stated in Craig and the *997 statutory standard under the provision of the Iowa Code upon which the trial court relied. Rather, it is readily apparent that the pertinent portions of the statute mirror the constitutional requirements set forth in Craig. Compare Craig, 497 U.S. at 855-56, 110 S.Ct. 3157 (the procedure is necessary only if a case-specific finding is made that the child witness will suffer from trauma caused by testifying in the physical presence of the defendant such that, at a minimum, the trauma would impair the child’s ability to communicate), with Iowa Code § 910A.14(1) (now § 915.38(1)) (“a court may protect a minor, as defined in section 599.1, from trauma caused by testifying in the physical presence of the defendant where it would impair the minor’s ability to communicate .... However, such an order shall be entered only upon a specific finding by the court that such measures are necessary to protect the minor from trauma.”). Moreover, the statute appears, on its face, to provide other means of ensuring that the testimony admitted against the accused “is reliable and subject to the rigorous adversarial testing that is the norm of Anglo-American criminal proceedings,” even in the absence of face-to-face confrontation, see Craig, 497 U.S. at 846-47, 110 S.Ct. 3157, because it provides, inter alia, that “[t]he judge shall inform the minor that the defendant will not be present in the room in which the minor will be testifying but that the defendant will be viewing the minor’s testimony through closed circuit television”; the statute permits cross-examination by defense counsel; and “[d]ur-ing the minor’s testimony the defendant shall remain in the courtroom and shall be allowed to communicate with the defendant’s counsel in the room where the minor is testifying by an appropriate electronic method.” Iowa Code § 910A.14(1) (now codified at Iowa Code § 915.38(1)). Nor does the decision of the Eighth Circuit Court of Appeals in Hoversten v. Iowa, 998 F.2d 614 (8th Cir.1993), demonstrate that any “unreasonable application” occurred in this case. In Hoversten, “[t]he trial court granted the motion [for testimony by a child witness behind a one-way mirror] without a hearing,” Hoversten, 998 F.2d at 615, but a pre-trial evidentiary hearing on the question of testifying outside of the presence of the defendant occurred in this case. Moreover, on habeas review in Hoversten, the Eighth Circuit Court of Appeals concluded that “[t]he Iowa trial court’s finding of necessity was based upon the ‘traumatic experience of testifying in open court,’ a consideration expressly held in Craig to be insufficient,” and “[t]he Iowa Supreme Court’s de novo review of the record likewise focused upon this abused child’s need for protection generally, rather than her specific need for protection from the experience of testifying in front of Hoversten.” Id. at 616. The Eighth Circuit Court of Appeals also rejected the notion that it should review the record to determine whether it supported a case-specific finding that no state court had made, which the court doubted Craig would permit it to do. Id. The court, therefore, held that the record did not show whether the child witness would be traumatized, not by the courtroom generally, but by the presence of the defendant. Id. This failing in the application of Craig also is not present here, because the trial court specifically found that testifying in Lomholt’s presence, not just testifying in a courtroom generally, would cause the children trauma that would impair their ability to communicate. Similarly, Lomholt did not challenge the counselor’s qualifications as an expert at the trial level and does not appear to do so now, nor did he or does he now challenge the fact that the only evidence presented at the pre-trial hearing was from such an expert. Rather, his trial counsel challenged the counselor’s bias and the suffi *998 ciency of the factual basis for her opinions. The Supreme Court indicated in Craig that a sufficient finding of necessity can be made, for example, “on the basis of the expert testimony before [the court],” without questioning the child witnesses in the defendant’s presence. Craig, 497 U.S. at 860, 110 S.Ct. 3157. Thus, the fact that the trial court made its determination here solely on the basis of the counselor’s testimony is not necessarily fatal to the “reasonableness” of its application of federal law as embodied in the Supreme Court’s decision in Craig. Moreover, the court has already concluded that the evidence before the trial court was sufficient that the trial court reasonably determined “that testimony of the child witnesses in the courtroom in the defendant's presence ‘will result in [each] child suffering serious emotional distress such that the child cannot reasonably communicate,’ ” thus satisfying factual predicates of the constitutional standard stated in Craig. Id. (quoting the Maryland statute). What is at issue, then, on the “unreasonable application” prong of the analysis in this case is whether the counselor’s testimony was sufficiently “case-specific.” See id. at 858, 110 S.Ct. 3157. Both Lomholt’s trial and habeas counsel contend that Tom-son’s testimony that all four- or five-year-old children would suffer trauma from testifying in front of someone who had abused them necessarily establishes that Tomson did not make a “case-specific finding” of “trauma,” but instead offered “expert” testimony based solely on her particular bias, such that the trial court unreasonably applied the Craig standards to the facts in this case. They make the same argument that Tomson improperly relied on studies of other children in formulating her opinions, rather than case-specific observations. This court cannot agree. Rather, the “all children” statement came at the close of Tomson’s direct examination, and could reasonably have been taken by the trial court as a demonstration that the case-specific observations and conclusions that Tomson had previously made about B.G. and N.P. in the course of her testimony were also consistent with conclusions drawn from more general studies, ie., that B.G. and N.P. were not having some feigned or aberrant reaction. To put it another way, the fact that case-specific observations and conclusions are consistent with an expert’s general opinions or other studies concerning the reaction of young children to testifying about wrongdoing in the presence of the perpetrator doesn’t mean that the expert’s opinion concerning the specific children in question wasn’t “case-specific.” For the same reasons, the court cannot embrace Judge Zoss’s suggestion that Tomson’s testimony that all four- or five-year-old children would suffer trauma from testifying in front of someone that had abused them was “the crux” of her testimony. See Report and Recommendation at 20. Rather, the bulk of Tomson’s testimony concerned specific observations and opinions about B.G. and N.P.; the fact that those observations and opinions were informed by Tomson’s experience, training, and knowledge of pertinent research does no more than demonstrate that Tom-son did what an expert should be expected to do in formulating an opinion. Judge Zoss concluded that there was “nothing in the opinion of the Iowa Court of Appeals to salvage the ruling of the trial court” — which this court concludes did not need salvaging — because he believed that the state appellate court had offered its observations that other evidence presented at trial to establish Lomholt’s guilt “collectively corroborate^] Lomholt’s criminal acts,” see Appendix of Documents, Exhibit B at 5, “as if this evidence somehow related to the case-specific showing required by Craig that the one-way closed-circuit tele *999 vision procedure was necessary to protect the welfare of the child witnesses.” Report and Recommendation at 21. This court does not read the findings of the Iowa Court of Appeals concerning corroboration of Lomholt’s criminal acts as having anything to do with ease-specific determinations of necessity of testimony by closed-circuit television. Rather, as the portions of the appellate decision quoted above demonstrate, the Iowa Court of Appeals reviewed the evidence before the trial court concerning B.G. and N.P., as provided by Tomson, before affirming the trial court’s case-specific determination, stating, “We find there was specific information presented to the satisfaction of the court to indicate the girls’ ability to testify would be impaired if not shielded from Lomholt.” See Appendix of Documents, Exhibit B at 4-5. The last paragraph of the opinion of the Iowa Court of Appeals, which addresses other evidence corroborating Lomholt’s criminal acts, deals with Lomholt’s confession, “independent of the girls’ testimony.” See id. at 5. Thus, the paragraph must be read as a conclusion that Lom-holt’s confession was sufficiently corroborated by other evidence to sustain his conviction, i.e., that any error in permitting the children to testify by closed-circuit television was “harmless.” Upon de novo review, the court finds no “unreasonable application” of federal law in the state trial court’s determination that the child witnesses should be allowed to testify by closed-circuit television or in the state appellate court’s affirmance of that determination. Therefore, the court must sustain the respondent’s second objection and reject Judge Zoss’s conclusion on the “unreasonable application” issue. 4. “Harmless error” Even if the court could accept Judge Zoss’s conclusions that the trial court either unreasonably determined the facts from the record or unreasonably applied federal law to the facts in allowing the children to testify by closed-circuit television, the court would be required to reject Judge Zoss’s conclusion that the Confrontation Clause error was not “harmless.” As Judge Zoss correctly noted, in Hoversten, the Eighth Circuit Court of Appeals explained the standard for “harmless error” analysis in the circumstances presented here: [In] Coy [the United States Supreme Court] held that, in analyzing the evidence for harmless error, the child’s testimony must be entirely excluded because it would be “pure speculation” to consider whether the child’s testimony, or the jury’s assessment of that testimony, would have changed had there been proper confrontation. “[H]armlessness must therefore be determined on the-basis of the remaining evidence.” 487 U.S. at 1022, 108 S.Ct. at 2803. Hoversten, 998 F.2d at 617. Although this court agrees with Judge Zoss that Hover-sten states the correct standard for “harmless error” review, this court reaches a conclusion different from Judge Zoss’s on the result of a “harmless error” analysis. If the children’s testimony is excluded from the “harmless error” analysis, then the evidence of Lomholt’s guilt consists primarily of his own confession. Judge Zoss correctly noted that both state and federal law require that a confession of the accused must be corroborated to sustain a conviction. See Iowa Code § 813.2, Rule 20(4); State v. Capper, 539 N.W.2d 361, 364 (Iowa 1995); Wong Sun v. United States, 371 U.S. 471, 488-89, 83 S.Ct. 407, 9 L.Ed.2d 441 (1963). However, the court cannot agree with Judge Zoss that “even if Lomholt made a totally uncoerced confession, nevertheless, there was no corroborating evidence that the crime occurred at all beyond the children’s testimony, which must be excluded as ex *1000 plained above, and hearsay testimony by the children’s mothers of what the children said occurred.” Report and Recommendation at 27. Rather, the Iowa Court of Appeals identified sufficient evidence to corroborate Lomholt’s confession, which consisted of evidence plainly demonstrating that Lomholt had the opportunity to commit the crime, and N.P.’s mother’s testimony that N.P.’s personality had changed after spending the evening with Lomholt. See Appendix of Decisions, Exhibit B at 5. The latter evidence was not a “hearsay” repetition of statements that N.P. made about touching by Lomholt, but N.P.’s mother’s own declaration about what she observed directly about N.P.’s behavior after the alleged abuse. See Fed. R.Evid. 801(c) (hearsay is a statement other than one made by the declarant); and •compare Hoversten, 998 F.2d at 617 (the purportedly corroborating evidence was indeed hearsay, consisting of the testimony by two family members, a police investigator, and a physician “as to statements the child had made about Hovesten touching her”). The contention of Lomholt’s habeas counsel that this statement was nevertheless “hearsay,” because N.P. might have been trying to make a non-verbal statement about Lomholt by her behavior, is too strained to countenance. However, even if N.P. had somehow been making a “statement” by her behavior, it was a statement of her then existing state of mind, emotion, sensation, or physical condition, which is excepted from the hearsay rule, see Fed.R.Evid. 803(3), not a statement about what Lomholt did to her, as in Hoversten. Consequently, any error in permitting the children to testify by closed-circuit television in this case was either “harmless beyond a reasonable doubt” or had no “substantial and injurious effect or influence in determining the jury’s verdict,” where Lomholt’s confession was sufficiently corroborated by other, admissible evidence in the record. Cf. Hoversten, 998 F.2d at 617 (citing Chapman v. California, 386 U.S. 18, 24, 87 S.Ct. 824, 17 L.Ed.2d 705 (1967), for the first standard and Brecht v. Abrahamson, 507 U.S. 619, 113 S.Ct. 1710, 123 L.Ed.2d 353 (1993), for the second, and concluded that the court need not reach the question of which standard was applicable, because “the error in this case was not harmless under either standard”) (emphasis added). Therefore, the court sustains the respondent’s third objection to the February 5, 2002, Report and Recommendation, and rejects Judge Zoss’s contention that the Confrontation Clause error was not “harmless.” D. Certificate Of Appealability Because Judge Zoss recommended that relief be granted on Lomholt’s petition, he did not consider whether or not a certificate of appealability should issue in the event the petition was denied. In order to obtain a certificate of appealability to appeal the denial of his petition for habeas corpus relief, Lomholt must make a substantial showing of the denial of a constitutional right. See 28 U.S.C. § 2253(c); see also Garrett v. United States, 211 F.3d 1075, 1076-77 (8th Cir.), cert. denied, 581 U.S. 908, 121 S.Ct. 254, 148 L.Ed.2d 184 (2000); Mills v. Norris, 187 F.3d 881, 882 n. 1 (8th Cir.1999); Carter v. Hopkins, 151 F.3d 872, 873-74 (8th Cir.), cert. denied, 525 U.S. 1007, 119 S.Ct. 524, 142 L.Ed.2d 435 (1998); Ramsey v. Bowersox, 149 F.3d 749 (8th Cir.1998), cert. denied, 525 U.S. 1166, 119 S.Ct. 1083, 143 L.Ed.2d 85 (1999); Cox v. Norris, 133 F.3d 565, 569 (8th Cir.1997), cert. denied, 525 U.S. 834, 119 S.Ct. 89, 142 L.Ed.2d 70 (1998). As the Eighth Circuit Court of Appeals has explained, “ ‘A substantial showing is a showing that issues are debatable among reasonable jurists, a court *1001 could resolve the issues differently, or the issues deserve further proceedings.’ ” Garrett, 211 F.3d at 1077 (quoting Cox, 133 F.3d at 569). The court finds that Lom-holt has made the necessary showing here for a certificate of appealability to issue in this case, not least because of the difference of opinion on the key issues between Judge Zoss and the undersigned. III. CONCLUSION The court holds that the state trial court made neither an “unreasonable determination” of the facts in light of record evidence, nor an “unreasonable application” of federal law, as embodied in Maryland v. Craig, 497 U.S. 836, 110 S.Ct. 3157, 111 L.Ed.2d 666 (1990), to the facts, in permitting two child victims to testify by closed-circuit television, outside of Lomholt’s presence, in Lomholt’s trial on sex abuse charges. Even assuming that there was a Confrontation Clause error, the court concludes that such error was “harmless” as a matter of law in light of sufficient admissible evidence to corroborate Lomholt’s confession. THEREFORE, 1. The court sustains the respondent’s objections to the February 5, 2002, Report and Recommendation, and rejects the challenged portions of the Report and Recommendation. 2. Upon de novo review, Lomholt’s petition for habeas corpus relief is denied. Judgment shall enter in favor of the respondent. 3. A certificate of appealability pursuant to 28 U.S.C. § 2253(c) shall issue for Lomholt to appeal this disposition of his petition, should he choose to do so. IT IS SO ORDERED. Footnotes [1] Although no party contends that the Iowa statute authorizing the use of testimony of child witnesses via closed-circuit television fails on its face to meet the applicable constitutional standard, it may nevertheless be instructive to know what the statute provides in its entirety. The statute cited by the trial court in Lomholt's case as the authority for the procedure used to receive the testimony of the child witnesses at trial provides as follows: 1. Upon its own motion or upon motion of any party, a court may protect a minor, as defined in section 599.1, from trauma caused by testifying in the physical presence of the defendant where it would impair the minor’s ability to communicate, by ordering that the testimony of the minor be taken in a room other than the courtroom and be televised by closed circuit equipment for viewing in the courtroom. However, such an order shall be entered only upon a specific finding by the court that such measures are necessary to protect the minor from trauma. Only the judge, prosecuting attorney, defendant's attorney, persons necessary to operate the equipment, and any person whose presence, in the opinion of the court, would contribute to the welfare and well-being of the minor may be present in the room with the minor during the minor's testimony. The judge shall inform the minor that the defendant will not be present in the room in which the minor will be testifying but that the defendant will be viewing the minor’s testimony through closed circuit television. During the minor's testimony the defendant shall remain in the courtroom and shall be allowed to communicate with the defendant’s counsel in the room where the minor is testifying by an appropriate electronic method. Iowa Code § 910A.14(1) (now codified at Iowa Code § 915.38(1)) (emphasis added). [2] As this court reads Craig, the showing that the trauma must be more than de minimis is not separate from the showing that the trauma must be such that it impairs the child’s ability to communicate. Rather, as this court reads Craig, a showing that the trauma impairs the child’s ability to communicate satisfies the requirement that "the emotional distress suffered by the child witness in the presence of the defendant is more than de minimis.” See Craig, 497 U.S. at 857, 110 S.Ct. 3157.