concurring in part and dissenting in part.
I agree that the proper test to be used to determine if the rule established by the United States Supreme Court in McKoy v. North Carolina, 494 U.S. 433, 108 L. Ed. 2d 369 (1990), should be applied retroactively is the test set forth in Teague v. Lane, 489 U.S. 288, 103 L. Ed. 2d 334 (1989). However, I do not believe that the rule set forth in McKoy satisfies the second narrow exception of Teague, which would require retroactive relief of McKoy error on collateral review.
“Under Teague, new rules may be applied . . . only if they come within ‘one of two narrow exceptions.’ ” Sawyer v. Smith, 497 U.S. 227, 241, 111 L. Ed. 2d 193, 211 (1990) (quoting Saffle v. Parks, 494 U.S. 484, 486, 108 L. Ed. 2d 415, 423 (1990)). “The second Teague exception applies to new ‘watershed rules of criminal procedure’ that are necessary to the fundamental fairness of the criminal proceeding.” Id. at 241-42, 108 L. Ed. 2d at 211 (quoting Saffle v. Parks, 494 U.S. at 495, 108 L. Ed. 2d at 429).
Unlike the majority, I am not persuaded by the analysis of the Fourth Circuit in Williams v. Dixon, 961 F.2d 448, cert. denied, --- U.S. ---, 121 L. Ed. 2d 445 (1992). In Williams, the court found that the rule set out in McKoy was a “ ‘bedrock procedural element[ ],’ ” id. (quoting Sawyer v. Smith, 497 U.S. at 242, 111 L. Ed. 2d at 211), “implicit in the concept of ordered liberty,” id. at 456, in part because the procedure had been described by the United States Supreme Court as “ ‘arbitrary or capricious,’ ” id. (quoting McKoy v. North Carolina, 494 U.S. at 454, 108 L. Ed. 2d at 387 (Kennedy, J., concurring)), . and did not provide the “ ‘fundamental respect for humanity underlying the Eighth Amendment,’ ” id. (quoting Woodson v. North Carolina, 428 U.S. 280, 304, 49 L. Ed. 2d 944, 961 (1976)).
The United States Supreme Court in determining the case of Caldwell v. Mississippi, found prejudicial error in a prosecutor’s comments which led a jury to the false belief that the responsibility *516for determining the appropriateness of the defendant’s capital sentencing rests elsewhere. 472 U.S. 320, 328-29, 86 L. Ed. 2d 231, 239 (1985). One member of the Court noted that such prosecutorial error created an “unacceptable risk that ‘the death penalty [may have been] meted out arbitrarily or capriciously.'” Id. at 343, 86 L. Ed. 2d at 248-49 (O’Conner, J., concurring in part and dissenting in part) (quoting California v. Ramos, 463 U.S. 992, 999, 77 L. Ed. 2d 1171, 1179 (1983)) (emphasis added). The Court held that Caldwell error might produce “substantial unreliability as well as bias in favor of death sentences.” Id. at 330, 86 L. Ed. 2d at 240. In spite of this language, the United States Supreme Court determined that Caldwell would not be applied retroactively to cases on collateral review, specifically finding that it was a “new rule” that did not satisfy the second exception of Teague. Sawyer v. Smith, 497 U.S. 227, 245, 111 L. Ed. 2d 193, 213 (1990).
Just as the rule set forth in Caldwell was not applied retroactively, neither should the rule set forth in McKoy be applied retroactively. I do not believe that retroactive application of the McKoy rule is a prerequisite to “fundamental fairness” of the type that comes within Teague’s second exception. See Wilcher v. Hargett, 978 F.2d 872 (1992) (determining that the McKoy rule was a new rule that would not be applied retroactively under the Supreme Court rules as set forth in Teague), cert. denied, --- U.S. ---, 126 L. Ed. 2d 63 (1993).
Justice Harlan first set forth the language used in Teague in his separate opinion concurring in part and dissenting in part in Mackey v. United States, 401 U.S. 667, 675, 28 L. Ed. 2d 404, 410 (1971). In Mackey, Justice Harlan noted that he believed that a new rule that should be applied retroactively would be one such as the right to counsel, which is now “a necessary condition precedent to any conviction for a serious crime.” Id. at 694, 28 L. Ed. 2d at 421 (Harlan, J., concurring in part and dissenting in part).
Finally, I am persuaded that McKoy error cannot at the same time be both subject to harmless error analysis (as we have held numerous times) and its retroactive effect be necessary to “the fundamental fairness of the criminal proceeding.” I note that this Court has found the failure to follow McKoy to be harmless error on five occasions: State v. Price, 334 N.C. 615, 433 S.E.2d 746 (1993); State v. Allen, 331 N.C. 746, 417 S.E.2d 227, cert. denied, *517--- U.S. ---, 122 L. Ed. 2d 775 (1992), reh’g denied, --- U.S. ---, 123 L. Ed. 2d 503 (1993); State v. Hunt, 330 N.C. 501, 411 S.E. 2d 806, cert. denied, --- U.S. ---, 120 L. Ed. 2d 913 (1992); State v. Laws, 328 N.C. 550, 402 S.E.2d 573, cert. denied, --- U.S. ---, 116 L. Ed. 2d 174, reh’g denied, --- U.S. ---, 116 L. Ed. 2d 648 (1991); State v. Roper, 328 N.C. 337, 402 S.E.2d 600, cert. denied, --- U.S. ---, 116 L. Ed. 2d 232 (1991). Thus, it is clear that jury instructions free of McKoy error are not “a necessary condition precedent to any conviction for a serious crime.” See Mackey, 401 U.S. at 694, 28 L. Ed. 2d at 421 (Harlan, J., concurring in part and dissenting in part). I believe that it is inconsistent to find that a right is so fundamental to the accuracy of the criminal proceeding as to require it to be applied retroactively but also find that a violation of this right is subject to “harmless error” analysis.
I would affirm the decision of Judge Albright, refusing to give McKoy retroactive relief and denying defendant’s motion for appropriate relief.