Gary Graham (a black man) was executed in Texas for murder. He was convicted primarily on the eyewitness testimony of one witness, who said she saw him commit the murder (at a distance of 30-40 feet) at night in the lighted parking lot of a grocery store. The eyewitness also said Graham passed within about 10 feet of the front of her car, and that she again identified him by the illumination of her headlights. My letter was in response to an article "Guilty as Charged," which appeared in the June, 28, 2000 issue of The Wall Street Journal.

I was reminded of the case of Lenell Geter (a black man) who was convicted and sent to prison for the robbery of a Kentucky Fried Chicken restaurant - based primarily on an eyewitness account. He was later cleared of the charges. I wrote an editorial in April 3, 1984 in The Dallas Morning News about the frailty of eyewitness testimony.

There is another problem with identifying black people, or any dark-skinned person, in poor lighting conditions. The light does not reflect well from their faces. If you watched CBS's "Survivor" and saw Gervase, the black contestant, in the light of torches, he was hard to identify - compared to the white faces sitting around him. Yeah, sure. You knew it was him because he was trapped on an island. Try it in a poorly-lighted parking lot in Baltimore at midnight. TV cameramen know about this at football games. When players run from the sunlight into a shaded area, such as often happens with the open roof at Texas Stadium, the facial features of the black players tend to disappear. The cameraman will "open" the lens and let more light in to get the black faces back. If you have ever used a light meter to take a group photo of whites and blacks together and used an "exposure meter," you have to work off the face of the black people to get a decent shot of the group. So, I am even more suspicious of eyewitness accounts regarding black, or other dark-skinned, defendants when they were seen at night in adverse lighting conditions.

A current study, published in the August issue of "NeuroReport," involved testing people to identify black and white faces. The bottom line seems to be that white folks tend to be able to identify white faces better than they could black, and vice versa. I think that is what they were getting at with all the mumbo jumbo. A specific part of the brain appears to react differently to faces of another race than to faces of the same race, according to the team of researchers.

All this goes along with my statement in the WSJ letter about cross-cultural identifications being the most troubling.

Back to a List of Topics in this Section

Back to Richard's Home Page

Richard Rhodes

07/04/2000