dissenting.
I cannot join the majority opinion for two reasons. First, I do not agree that the Confrontation Clause applied to the second phase of Lindh’s trial. On this issue, the majority adopts the analysis of Judge Wood, see Lindh v. Murphy, 96 F.3d 856, 879-83 (7th Cir.1996) (Wood, J., concurring in part and dissenting in part), who suggests that the second phase of Wisconsin’s proceedings are more akin to an ordinary trial’s determination of guilt than to a sentencing proceeding’s disposition of the defendant. I, however, remain unconvinced that this second phase is so different from non-capital sentencing proceedings, where neither we nor the Supreme Court have held that the Confrontation Clause applies.
Simply put, the stakes for a defendant are quite different at phase two of Wisconsin’s bifurcated proceedings than they are at
phase one. We must not forget that the jury found Lindh guilty in the trial’s first phase and that this verdict will remain with Lindh for the rest of his life regardless of any jury finding regarding mental disease. As the Wisconsin Supreme Court has described it, “[T]he question of guilt (i.e., was the conduct of the defendant criminal?) is settled in the felony prosecution that takes place in phase one.” State v. Koput, 142 Wis.2d 370, 418 N.W.2d 804, 812 (1988). When deciding the question of mental disease, by contrast, “we are largely concerning ourselves with the difference in the institutional treatment of the defendant.” Id. at 813 (quoting State v. Shoffner, 31 Wis.2d 412, 143 N.W.2d 458, 474 (1966) (Wilkie, J., concurring)). Had the jury found Lindh not guilty by reason of mental disease or defect, he still would have faced immediate confinement, possibly for life. See Wis. Stat. § 971.17(1). Although Judge Wood minimizes the severity of this punishment based on the possibility (and I emphasize possibility) of conditional release after six months, see Lindh, 96 F.3d at 882, such release is hardly the same as an acquittal. For at least six months after the termination of the initial confinement, the conditional release is revokable. See Wis. Stat. § 971.17(3)(e), (5). In short, the second phase was dispositional in nature, and we should be leery of letting Confrontation Clause rights creep into dispositional proceedings.
The second reason I cannot join the majority is that even if the Confrontation Clause did apply to the second phase of Lindh’s trial, the trial judge did not violate the Constitution by limiting the cross-examination of Dr. Roberts. If Lindh wanted to inquire into the sexual allegations to show Roberts’ bias, we have no evidence to suggest that Roberts ever changed his opinion to favor the State. The majority says Roberts’ consistency is not quite good enough because an unbiased Roberts might have changed his opinion in favor of Lindh. This argument, however, is quite speculative, and it becomes even more tenuous when considered in light of the underlying conjecture that Roberts would shade his *903testimony in Dane County so that the Milwaukee County prosecutor might cut him a deal. As we stated in our previous en banc decision in this ease, “if Roberts thought that, he probably also believed that his many prior engagements as a prosecutorial expert would carry even more weight.” Lindh, 96 F.3d at 877.
If, on the other hand, Lindh wanted to inquire into the sexual allegations to show that Roberts was not competent, such questioning was properly excluded. Roberts’ alleged misconduct was entirely unrelated to his diagnosis of Lindh and did not impeach his overall professional credibility. See State v. Lindh, 161 Wis.2d 324, 468 N.W.2d 168, 181 (1991). Judge Wood has argued that “[i]t is fanciful to assume that a jury would give the same weight to the ‘mid-west psychiatrist of the year’ as it would give to someone who was facing both criminal charges and civil proceedings to revoke his license.” Lindh, 96 F.3d at 884. But it is precisely because a jury might be swayed by this misconduct that it was proper for the trial judge to exclude the cross-examination. Although psychiatry is admittedly not a hard science, it has to be more than just personal vouching or it never should be admitted as expert testimony in the first place. The mentioning at trial that Roberts was a grandfather therefore did not open his psychiatric testimony to impeachment based on sexual misconduct and whatever other skeletons the defense could find in Roberts’ closet.
The Supreme Court has told us that “trial judges retain wide latitude insofar as the Confrontation Clause is concerned to impose reasonable limits on such cross-examination based on concerns about, among other things, harassment, prejudice, confusion of the issues ... or interrogation that is repetitive or only marginally relevant.” Delaware v. Van Arsdall, 475 U.S. 673, 679, 106 S.Ct. 1431, 1435, 89 L.Ed.2d 674 (1986). Although the aedpa does not apply to this case and thus does not add an extra layer of deference, see Lindh v. Murphy, — U.S. -, 117 S.Ct. 2059, 138 L.Ed.2d 481 (1997), a decision to limit cross-examination is still a discretionary decision in the hands of a trial judge. See, e.g., United States v. Dillard, 43 F.3d 299, 305 (7th Cir.1994). The trial judge here was faced with a defendant who wanted to pursue a potentially-inflammatory line of questioning regarding sexual misconduct. Although Lindh had the chance to press Roberts regarding whether his psychiatric opinion had changed, Lindh insisted on pursuing the more lascivious inquiry into the sexual allegations. The trial judge did not abuse his discretion by halting that inquiry. Moreover, I cannot agree with the majority’s suggestion that the trial judge was obligated to come up with his own compromise solution such as redaction.
Even if one believes that it is a close call regarding whether the trial judge here abused his discretion, it is nonetheless a call we have already made. In our original en banc decision, we stated that under the circumstances of this case,
a reasonable and responsible judge might conclude that cross-examination based on the allegations that Roberts had sexual relations with three female patients would divert attention from the principal issue at hand — whether Lindh had a mental disease in January 1988 — without producing concrete evidence of bias. When a district judge comes to such a conclusion, a federal court of appeals will exercise deferential review____ Section 2254(d)(1) [after amendment by the aedpa] requires a federal court hearing a collateral attack to accord at least that much respect to the state courts.
Lindh, 96 F.3d at 877. Thus, before we applied the aedpa’s new standard of review, we stated that the decision to limit cross-examination could have been made by “a reasonable and responsible judge.” That hardly sounds like an abuse of discretion to me.
Finally, I would note that the harmless-error standard for this case is not as strict as it would be for a case coming to us on direct appeal. In a habeas ease involving constitutional trial error, the error must have result*904ed in actual prejudice-meaning that the error “had substantial and injurious effect or influence in determining the jury’s verdict” — before a writ of habeas corpus may be granted. Brecht v. Abrahamson, 507 U.S. 619, 637, 113 S.Ct. 1710, 1721, 123 L.Ed.2d 353 (1993) (quoting Kotteakos v. United States, 328 U.S. 750, 776, 66 S.Ct. 1239, 1253, 90 L.Ed. 1557 (1946)). Confrontation Clause cases are subject to harmless error review, see Coy v. Iowa, 487 U.S. 1012, 1021, 108 S.Ct. 2798, 2803, 101 L.Ed.2d 857 (1988), and any error in this case was harmless under the Brecht standard. It was Lindh, after all, who had the burden of proof at the second phase of the trial. See Wis. Stat. § 971.15(3). The State, moreover, had a second psychiatric expert besides Roberts who testified that Lindh “did not suffer a brief reactive psychosis at the time of the crimes, did not have a mental disease, and suffered no impairment of his ability to conform his conduct to the requirements of the law.” Lindh, 468 N.W.2d at 175. Even if one discounts all of Roberts’ testimony as biased and unreliable, Lindh would have had a tough time in the face of the testimony of the State’s other expert. I therefore do not understand how the majority can conclude that the limitation on Roberts’ cross-examination had, in the words of Brecht, “substantial and injurious effect or influence in determining the jury’s verdict.” Cf Gomez v. Ahitow, 29 F.3d 1128, 1137 (7th Cir.1994) (finding no actual prejudice in habeas ease involving Confrontation Clause).
For the foregoing reasons, I respectfully dissent.