Cortez, Raul ( 2011 )

Death Opinion













IN THE COURT OF CRIMINAL APPEALS OF TEXAS

---

NO. AP-76,101  

---

RAUL CORTEZ, Appellant

v.

THE STATE OF TEXAS

---

ON DIRECT APPEAL FROM COLLIN COUNTY

---

   Johnson, J., filed a concurring and dissenting opinion.



C O N C U R R I N G A N D D I S S E N T I N G O P I N I O N

DNA evidence is coming to be regarded as absolute proof of guilt, but given the current state of the science, that status is not yet earned and will not be until DNA experts learn how to testify about: (1) statistically valid DNA evidence in a way that; (2) accurately conveys the significance of the evidence; and (3) can be adequately understood by the jurors. Dr. Staub and the staff at Orchid Cellmark did an admirable job of differentiating between samples that were a good basis for comparison and those that were not. ⁽¹⁾ Dr. Staub clearly explained the process of DNA amplification and how samples are compared and matched. Problems appeared, however, when the evidence turned to frequency of occurrence of particular alleles in racial and ethnic groups.

Dr. Staub testified that "[t]o say a suspect matches an evidence sample, although you might think that means a lot, it really means nothing until you assign some sort of a probability that a random person wouldn't match that sample, because obviously if that is high, then the fact that the suspect matches is not quite so important as if it were an extremely rare match probability." XXXXII R.R. at 61. However, Dr. Staub's definition of random match is backwards: the actual standard for stating a probability is the probability that a randomly chosen person will match the DNA profile, not that the randomly chosen person will not match. Put simply, if I stand on a street corner and snag the first person who catches my eye, the probability that that person will match a previously selected DNA profile is governed by the probability of a random match.

My discomfort with Dr. Staub's estimates of probability grew with his testimony about how he had calculated those estimates. By definition, probabilities must be based on a large number of actual observations, yet Dr. Staub began by using Orchid Cellmark's own data base, which held only 1276 profiles of Caucasians and 597 profiles of Hispanics. Those data bases are much too small for accurate calculations of probability, especially in Texas, which has a high Hispanic population, and the data base has less than half as many samples for Hispanics than for Caucasians, who are a minority in Texas. Dr. Staub conceded as much when he stated that, "to get a really powerful assessment of very fair frequencies," i.e. valid probabilities, for DNA matches through the male line, Y-STR, "would require a huge data base of male profiles, which we just don't have, and probably never will have a data base big enough to give anywhere close to the probabilities that we can give with standard autosomal STRs . . .." XXXXII R.R. at 70-71.

Dr. Staub was on more stable ground when his testimony shifted to probability based on the FBI data base. One of my concerns about the statistics he used is the apparent assumption made by Dr. Staub and the prosecution that Cortez is a "Southwest Hispanic"--defined as being from the "southwestern United States and consists predominantly of Hispanic individuals from Mexico or other Central or South American countries," as opposed to Southeast Hispanic--defined as an individual from "Cuba and other islands in that vicinity." XXXXII R.R. at 71. There doesn't appear to have been any exploration of whether Cortez was, in fact, a Southwest Hispanic, or whether he or his ancestors came from a Caribbean island, the Americas, Spain, or any of the former Spanish colonies, such as the Philippines, where Spanish surnames are common, or a mixture of two or more such regions.

And what makes a person "Hispanic"? Is it language? Or geographical origin? Or ethnic origin? Or "looking Hispanic"? Or surname? See Wilson v. State, 185 S.W.3d 481, 486-91 (Tex. Crim. App. 2006) (Johnson, J., concurring). As a group, persons for whom Spanish is their first language have varying fractions of Native American, Caucasian, Asian, and African ancestry. One has only to look at professional baseball to observe that many Spanish speakers with "Hispanic" surnames have a substantial contribution from African ancestors. Do we then use Hispanic DNA data or African-American DNA data or African DNA data or some algorithm that takes two or more into account? A number of professional boxers, especially in the lighter weight categories, have "Hispanic" surnames but are Filipinos and may or may not speak Spanish and may or may not have any Spanish ancestry. One might even encounter a person with an "Hispanic" surname, and who "looks Hispanic," but whose ancestors came from Spanish Morocco or Spanish Sahara, areas occupied by Arabs, or Equatorial Guinea, a former Spanish colony stuffed between Cameroon and Gabon on the southwest coast of Africa. What DNA statistics are appropriate for a person whose mother is Caucasian and whose father had a Caucasian mother and a Mexican father? The person's surname is "Hispanic," yet his ethnic background is three-quarters Caucasian. Were such questions even considered, much less asked and answered before the experts began spouting probabilities?

In this case, DNA was not the dominant evidence introduced against appellant during the guilt phase. I would affirm the verdict of guilty, but reverse on voir dire error that reversibly affected the punishment phase and order a new punishment hearing. Because the Court does not do so, I dissent.

Filed: September 14, 2011

Do not publish

1. Orchid Cellmark called some samples a match, but chose to defer on samples that fell below its threshold of a statistical match.