dissenting.
I respectfully dissent from the reversal decision and the majority’s reason for ordering it.
Admittedly, the Arizona Supreme Court has determined that evidence of DNA probability calculations is error. State v. Bible, 175 Ariz. 549, 584-87, 858 P.2d 1152, 1187-90 (1993), cert. denied, sub nom. Bible v. Arizona, — U.S. —, 114 S.Ct. 1578, 128 L.Ed.2d 221 (1994). But that admission is hardly, the end of the analysis. Traditionally, an evidentiary error at trial is harmless if the reviewing court can say beyond a reasonable doubt that the error did not contribute to the verdict. Bible, 175 Ariz. at 588, 858 P.2d at 1191. State v. Lundstrom, 161 Ariz. 141,150 & n. 11, 776 P.2d 1067, 1076 & n. 11 (1989); Chapman v. California, 386 U.S. 18, 24-26, 87 S.Ct. 824, 828-29, 17 L.Ed.2d 705 (1967) (the leading United States Supreme Court decision on harmless error).
The majority opinion rests wholly on Bible, which found that DNA probability evidence in that case was harmless error. While noting the great potential for prejudice from this type of error, Bible found its facts unusual because the DNA probability evidence was disputed and all the other evidence pointed to the one conclusion that the defendant committed the crime.
If the evidence against Defendant had been closely balanced, strong, or even very strong, we think it would be impossible to say beyond a reasonable doubt that the inadmissible DNA evidence did not affect the verdict. Evidence of odds even as “low” as one in sixty million that the blood on Defendant’s shirt was not the victims’s blood is, to say the least, powerful. Factually, however, this is a very unusual case. Given the testimony describing the dispute over the DNA evidence, that evidence was far from the most telling part of the State’s ease. The other evidence points with unerring consistency to one inarguable conclusion: that Defendant killed the victim.
Bible, 175 Ariz. at 588, 858 P.2d at 1191.
The majority in the present case apparently believes that anything other than that set of facts coupled with DNA probability evidence constitutes reversible error. The majority reads the quoted paragraph to mean that the DNA probability evidence is harmful error unless it is disputed and there is additional “far beyond overwhelming” evidence. I disagree with this reading.
While Bible recognizes that DNA probability evidence is potentially very prejudicial, it does not demand the unusually high burden the majority suggests. Bible does not announce a new test, nor does it change the traditional harmless error test.
First, nothing in Bible expressly alters the traditional harmless error test. In fact, that test as stated in Lundstrom is cited with approval. Id. In addition, Bible notes that the state has the burden of convincing the reviewing court that error is harmless, citing Chapman, 386 U.S. at 24-26, 87 S.Ct. at 828-29, and that the error is to be considered in light of all the evidence. Bible, 175 Ariz. at 588, 858 P.2d at 1191.
Second, Bible misinterprets Sullivan v. Louisiana, 508 U.S. -, ---, 113 S.Ct. 2078, 2081-82, 124 L.Ed.2d 182 (1993) (see majority opinion and Bible, 175 Ariz. at *46588, 590, 858 P.2d at 1191, 1193). In Sullivan, the United States Supreme Court merely discussed how an improper reasonable doubt jury instruction would always be reversible error. Sullivan, 508 U.S. at -, 113 S.Ct. at 2082. Sullivan does discuss Chapman and the traditional harmless error test briefly, but that test is never applied nor changed. Id. 508 U.S. at -, 113 S.Ct. at 2081. The quotes in Bible taken from Sullivan discuss how an appellate court cannot substitute its judgment for that of the trier of fact when the guilty verdict may not have been beyond a reasonable doubt. Contrary to the majority’s view, the Sullivan excerpts in Bible do not reflect any change in harmless error analysis but only reflect the Court’s inability to apply that analysis to the wholly dissimilar facts in Sullivan whose references are misleading as applied to the facts of Bible and to the present case. Sullivan is a case about fundamental reversible error in a jury instruction, not about the principle of harmless error. Any reliance by the majority on Sullivan as a standard for harmless error review is without basis.
Third, Bible’s “closely balanced, strong or even very strong” language about the weight of the evidence is not inconsistent with the traditional test requiring a showing beyond a reasonable doubt that the error did not affect the verdict. The quoted language simply parallels the showing required to meet the different evidentiary burden. A preponderance of the evidence (“closely balanced”), or clear and convincing evidence (“strong or very strong”) may be insufficient, but beyond a reasonable doubt evidence (“overwhelming”) is sufficient to uphold a verdict under the harmless error test when there is independent evidence of guilt beyond a reasonable doubt.
While our supreme court views DNA probability evidence as “powerful,” there is no intimation in Bible that DNA probability evidence usurps the harmless error standard. The question then becomes whether the error in the present case is harmless.
The DNA statistical evidence was not, as the majority states, presented for “two days.” The DNA evidence which was properly admissible did consume about two days but the improper evidence of probability— the only error—came at the end of a seven-day trial and consumes about 10 transcript pages—some 20 minutes.
The majority distinguishes this case from Bible for the following reasons: (1) the error was not mitigated and (2) the remaining evidence was insufficient under the harmless error test.
First, Bible cites cases that have upheld guilty verdicts with “weaker, or at least comparably strong, evidence of guilt independent of the erroneous admission of DNA evidence.” Bible, 175 Ariz. at 589-90, 858 P.2d at 1192-93. In the four cases cited by Bible, (People v. Barney, 8 Cal.App.4th 798, 10 Cal.Rptr.2d 731 (1 Dist.1992); State v. Nielsen, 467 N.W.2d 615, 619 (Minn.1991); People v. Wallace, 14 Cal.App.4th 651, 17 Cal. Rptr.2d 721 (1 Dist.1993); People v. Atoigue, 1992 WL 245628 (D.Guam App.Div. Sept. 11, 1992)), only one recognizes any mitigation of the DNA statistical evidence, Wallace, 17 Cal.Rptr.2d at 725-26.
Second, while the harmless error analysis in Bible is misleading, I believe it does not change the traditional harmless error test. In my view, there is sufficient independent and properly admitted evidence in the record to support the guilty verdict.
Police seized a night stick with dried blood residue on it from beneath Clark’s mattress. Clark identified it as his. During a search of Clark’s apartment, police found a black scarf, a length of nylon rope, dark shorts with a white waistband, ski gloves and a pair of running shoes in his closet. The victim testified that the assailant wore ski gloves with a velcro strap and dark shorts with a white waistband, tied her up with a nylon or hemp rope, and beat and sexually assaulted her with a stick or club. She testified that the gloves, shorts, rope and nightstick found in Clark’s apartment were consistent with items used in the assault.
Blood of a type and gene consistent with the victim was found on the ski gloves, the running shoes, the rope and the scarf from Clark’s apartment. Blood of a type consistent with Clark, found in less than 3% of the Caucasian population, appeared on a vaginal *47swab of the victim taken after her assault. DNA samples consistent with the victim were found on the scarf and ski gloves. DNA samples consistent with Clark were found on the ski gloves.
A forensic anthropologist and a podiatrist testified that the running shoes had more likely been habitually worn by Clark, not by his roommate. Moreover, Clark’s roommate and his roommate’s boss testified that the roommate could not have committed the crime because the roommate was working as the cashier of a Tempe gas station from 10 p.m. to 6 a.m. the night of the assault.
While the victim could not positively identify Clark, her description of the assailant matched Clark. She also stated that she and her daughter had observed a neighbor fitting Clark’s description watching her shortly before the assault. The victim testified about her physical resistance; Clark had bruising and scratching on his arm that he could not explain.
Clark offered no defense. He did not testify. His only witness was a police officer who testified about an irrelevant issue concerning the victim’s movement toward her front door during the assault.
Even without the improper DNA statistics, the evidence of Clark’s guilt is overwhelming. As in Bible, the other evidence points directly to the one inescapable conclusion that Clark committed the crime. I am convinced beyond a reasonable doubt that the introduction of the DNA probability evidence did not affect this verdict. I would affirm the conviction and along with it the continuing viability of the harmless error standard notwithstanding erroneous DNA probability evidence.