A Jury of Your Peer

Reader Al (aka Guest Blogger Al) sent me an email this morning. He threw me a bone on my Documents Gone Wild post, then linked to an article he thought worthy of note.

John,  

Good post on the falsification of documents in PHIII's case. 

Check out this article below 

http://colorlines.com/archives/2012/04/study_racial_make-up_of_jury_pool_affects_trial_outcome.html

Al's link summarized the findings of an article in the most recent issue of The Quarterly Journal of Economics. Since no one wants to read such an egghead article in its entirety, or even a summary of an article, or even this summary of the summary of the article, I'm going to display a graphic from the summary that gets straight to the bottom line.

You may think that this reduction in conviction rate of black defendants stems from having blacks on the jury. As it turns out, that's not quite what they found. The eligible jury population in those counties was less than 5% black. When a jury pool contained at least one black person, it typically contained no more than one. And here's the kicker. The black person did not actually have to be selected as a juror to reduce the chance that a black defendant would be convicted.

Very interesting.

Given that I sometimes find very interesting things to be somewhat interesting, I tried to understand why such a relationship might exist, assuming that it does exist. Here's what the authors had to say:

Having established that the racial composition of the jury pool has a substantial impact on conviction rates, we consider a number of possible channels through which random variation in the composition of the jury pool might affect trial outcomes. Most obviously and directly, having at least one black member in the jury pool makes it feasible to have a black member on the seated jury. Black representation on the seated jury might affect trial outcomes not only through the jury deliberation and decision process but also by affecting how the case is presented and argued by the prosecution and defense attorneys. 

Adding black potential jurors to the pool can also affect trial outcomes even when these jurors are not ultimately seated on the jury. This indirect effect comes about through the jury selection process if attorneys on each side use their peremptory challenges to strike the potential jurors most likely to be hostile to their case.

The authors then explain (but not clearly though) how the attorneys might use their peremptory challenges to select or exclude the black person from the jury. The modified peremptory challenge strategies (they argue) cause both the defense and the prosecution to compose a jury less hostile to a black defendant.

It sounded to me pretty much like hog spittle. I wasn't ready to buy either their surprising conclusion or their explanation for their surprising conclusion. I was trying to poke a hole in either their study or their argument when I noticed that the average jury size in their study was 7.11 jurors.

What the hell?

That forced me to learn something new, yet again. (Will it never end?) You may or may not have noticed that the study was limited to two counties in Florida. You may or may not have known that Florida does not seat twelve people on its juries, not even on its criminal juries. It seats just six jurors, plus one or two alternates. Hence the 7.11 jurors, on average.

That sucks.

As my fingers danced furiously over my aging keyboard, I learned that the early part of the 1970's was a bad time for our jury-based justice system. In the 1970 case of Williams v. Florida, the Supreme Court of the United States of America narrowly ruled that the Bill of Rights did not demand twelve jurors, as common law had demanded for centuries. Six were just as good as twelve. Just as good. Absolutely just as good. No doubt about it.

[Uh, oh!]

Also in 1970, in the case of Johnson v. Louisiana, The Supremes decided that unanimous verdicts were also unnecessary.

[Gulp!]

And just in case anyone doubted their sanity, The Supremes reaffirmed their no-need-for-unanimity decision in 1972, via their decision in Apodaca v. Oregon. 

[Mama!]

Welcome to the infamous slippery slope. Just how many jurors must find you guilty before the State can frog-march you off to prison, or to the gurney? Can a state use just 3 jurors? If so, do the three have to be unanimous?

Can the state try you with just one juror, by a jury of your peer? If your peer has multiple personalities, do they have to be unanimous?

Fortunately, The Supremes deslipperized the slippery slope in the 1978 case of Ballew v. Georgia. Georgia had ventured too far down the slippery slope, even for The Supremes. Georgia had taken a shine to convicting people with just five jurors. The Supremes said "Don't be ridiculous." Actually, according to the link, what they said was more along the line of:

The Court found that a trial by jury of less than six members violated the accused's right to a trial by jury as protected by the Sixth and Fourteenth Amendments. Justice Blackmun reasoned that small juries foster poor group deliberation. Group memory of the details of testimony, the ease with which group compromises can be made, and the desire of the group to be self-critical and reflective are all hindered as the size of the jury decreases. Blackmun also relied on statistical studies to claim that the risk of jury error increased with smaller juries.

Since even a non-Supreme-Court-Justice such as I can see that the summary would read just as well if it said "a jury less than seven members", you might guess that the Supremes used the opportunity to correct their screw-ups in Williams v. Florida, and Johnson v. Louisiana, and Apodaca v. Oregon.

But noooooo. As The Jury Expert explains:

Despite acknowledging "that the purpose and functioning of the jury in a criminal trial is seriously impaired, and to a constitutional degree, by a reduction in size to below six members," the Court reaffirmed its decision in Williams.

I now make the tricky U-turn back to the original point of this article. I think the Duke study discussed earlier in this article may be adding to the body of empirical studies finding that small juries are not as likely to be just juries. Small juries such as those described in the Duke study are more likely to exclude minorities, and less likely to be self-critical and reflective.

Let me know what you think. Should we demand that juries are composed of twelve jurors and that the verdict be unanimous? If not, how many jurors should it take to convict or exonerate a defendant in a criminal trial?

The Case of Preston Hughes III: Documents Gone Wild

Reader Lando is the winner of the first ever Skeptical Juror Brain Teaser. I hereby award Reader Lando the 10 attaperson points I promised, and I immediately convert them to Skeptical Juror points. All future awards will be in Skeptical Juror points, and I will keep a running total. I may eventually convert the Skeptical Juror points to physical script, though I don't want to run afoul of federal banking laws.

Reader Lando recognized correctly that the Consent to Search form was a composite document formed by placing the text block of a Consent to Search form over the signature block of some other form. (More specifically, the text block was placed over the upper half of an HPD Form No. BT-0012.) When one constructs such a composite form, particularly one so cheesy as the HPD Consent to Search form, the form becomes slightly three-dimensional. Copiers frequently fail to conceal the edges of the upper layer.

I present once again the Consent to Search form below, this time placed alongside its sister Consent for Samples form. The annotations are mine. Click to enlarge.

If you look near the upper and lower edges of the red box on the left, you can see the edge marks that Reader Lando picked up on. (Wow! I ended that sentence with two prepositions.) The upper half of the Consent to Search Form is entirely consistent with the upper half of the Consent for Samples form. The two documents have similar title and date blocks. The wording in the text block has been changed only as necessary to authorize the search of a residence rather than a collection of samples. Both text blocks begin and end with the same legalese

"I, Preston Hughes, having been informed of my constitutional right not to have ..." 

"This written permission is given by me ... voluntarily without threats or promises of any kind and is given with my full and free consent."

The only substantive differences between the upper portions of the two forms are the edge marks that Reader Lando so attentively noticed.  Reader Lando, on the other hand, made no mention of the many differences on the lower portions of the two forms. In fact, it suddenly seems as if the two forms are not siblings. They're distant cousins, at best. The signature blocks are formatted differently, and one has an actual form number while the other does not.

Also, Sgt. Ross's handwriting seems to have undergone a remarkable transformation in the two hours since she allegedly signed those two documents. In the Consent for Search form, her handwriting (when writing the time) looks remarkably similar to that of Sgt. Bloyd. In the Consent for Samples form, she wrote the time so carelessly that one might suspect someone else wrote it for her.

In summary, it appears as if the fine folks at the Houston Police Department took a blank Consent to Search form, filled in the information they wanted, cut or tore the text block from the form, taped the text block over a different form that Preston Hughes had in fact signed, and then copied the menagerie. Their forgery, however, was no more professional than their investigation in this case, and evidence of the falsification has been preserved lo these many years.

 *** THE END ***

But wait! There's more.

I figured if they pasted that text block over the upper portion of Form No. BT-0012, they may not have aligned everything properly.

So I checked to see if the text and underlines are parallel, upper to lower halves. They are, within the accuracy expected photocopied and scanned images.

So I checked to see if the text columns are properly aligned, upper to lower halves. They are, within the accuracy expected of photocopied and scanned images.

So I checked to see if the lines are spaced consistently, upper to lower halves. They are not.

They are not.

Check it out. Using Gimp, the poor man's Photoshop, I created a layer over the text block and placed a red line under each line of text, thus:

Now I can move these red lines down and see if the typed text in the lower half was typed on the same sheet of paper on a single pass through a typewriter. (Recall that this was in 1988.) The only typed text in the lower half  (in the same font as the upper half) are the words Signed and WITNESSES. Here's what I got.

When I kept the upper text properly underlined, the lower text was not properly underlined. Something's amiss. So I moved the lines up just a bit, so that they properly underlined (Signed) and WITNESSES. Not surprisingly, given the previous image, that didn't work either.

There is no way I can get the upper text to be underlined properly at the same time I can get the lower text underlined properly, and vice versa. Something happened to throw the line spacing off between the upper and lower halves, and it happened just about where there seems to be an edge mark, just above Preston's signature.

The entire case against Preston Hughes III depends on the veracity of the Houston Police.

We must believe Sergeant Hamilton when he claims that Shandra provided him a dying declaration, though no one else heard that declaration, and though Shandra was statistically likely to have been unconscious.

We must believe Sergeant Gafford when he claims that he did not coerce the first confession, because Gafford chose not to record the interrogation.

We must believe Sergeants Ferguson and Yanchak when they claim they did not coerce a second, entirely different confession, because Ferguson and Yanchak chose not to record the confession.

We must believe that the HPD did not plant Shandra's eyeglasses (or somebody's eyeglasses) in Preston's apartment, though the Evidence Inventory form makes it clear the police re-entered Preston's apartment  before 2:58 AM, at least 2.5 hours before Preston allegedly signed the consent to search form.

If the Consent to Search Form was falsified, as it now so clear it was, then we would be simply foolish to take the HPD at their word on this case.

Now, having completed this brain teasing interlude, I shall return to plodding carefully through the data. For me it is unfortunately no surprise that the police sometime manufacture evidence. For some of you readers, I suspect it will come as a disillusionment.

 <-- Previous                          Table of Contents                               Next --> 

The Case of Preston Hughes III: Brain Teaser #1

We need to talk.

This case is going to take a long time to work through. It's going to take a lot of posts. I'm going to plod through lots of details. I might lose many or most of you along the way. So be it.

It's beginning to feel as if I'm in this one for the long haul.

I have a couple thoughts on how I might continue to plumb the depths of this case while not allowing this blog to become focused on only one case. I simply need to find the time to blog about other cases and other topics. I've had two emails from readers about relevant topics. One mentioned the imperfect nature of fingerprints. The other mentioned the alleged CSI effect on juror expectations. Good topics, both. I'll be writing on them.

I'm also preparing a supplement (not an errata, a supplement) to my letter to Governor Jay Nixon in support of Byron Case's application for absolute pardon. This supplement is in response to the criticisms of reader Ivan. I'll soon be posting that as well.

Finally, I want to engage you more directly. I want you, those of you who are willing, to feel a bit of what I feel when I examine evidence closely and find what everyone before has missed. I want to give you a taste of what it feels like to chip away at a seemingly rock-solid case of guilt and uncover a wrongful conviction house-of-cards.

I'm going to start engagement here, in this post. I'm going to present you with three police documents from the Preston Hughes case. One is pretty clearly falsified. The others have been fudged only a little. I suggest there is sufficient information in the documents themselves for you to figure out which is which. You actually need to know nothing about this case, though you can catch up if you wish by reviewing the ever-current, always-evolving Table of Contents.

To keep from providing any hints via the order in which I present the documents, I'll use a random number generator to decide their order. Then you can use the comments to reveal which document is probably false and which are merely fudged. Please explain your reasoning in lucid, brilliant prose. Sign your comment with a pseudonym so that others can heap praise upon your perspicuity.

The person with the best answer will receive 10 attaperson points.

Get ready. Here come the documents. Click on each to enlarge. You can ignore the non-standard aspect ratios.

#1: We'll call this document the Consent for Samples.

#2. We'll call this document the Consent for Search.

#3. We'll call this document the Evidence Invoice. The red markings are from the folks at the Preston Hughes blog. You can ignore them.

There you go. Everything you need. Tell me, which of the three is probably falsified? How can we prove that? For the other two documents, how have they been tampered?


ADDENDUM:
We have a winner. Read the comments. For my more thorough explanation of how one of the documents was forged, see Documents Gone Wild.

 <-- Previous                            Table of Contents                             Next --> 

The Impending Execution of Thomas Kemp

Thomas Kemp sits on death row awaiting execution by the people of Arizona. He is scheduled for execution on 26 April. He has declined to seek clemency from the Arizona Board of Executive Clemency since he considers the clemency process to be a sham. (On this point, I may agree with him.) It is unlikely that Thomas Arnold Kemp will survive the day.

I provide a summary of his case from the recent adverse decision in Kemp v. Ryan (2011).

Thomas Arnold Kemp raises three issues in his appeal from the district court's denial of his habeas petition seeking relief from his state conviction for felony first-degree murder, armed robbery and kidnapping and from his capital sentence. ... The underlying criminal acts were described as follows by the district court: 

On July 11, 1992, at approximately 11:15 p.m., Hector Juarez awoke when his fiancée, Jamie, returned from work to their residence at the Promontory Apartments in Tucson. A short time later, Juarez left to get something to eat. Jamie assumed he went to a nearby fast food restaurant. 

At around midnight, Jamie became concerned that Juarez had not come home and began to look for him. She found both her car and his car in the parking lot. Her car, which Juarez had been driving, was unlocked and smelled of fast food; the insurance papers had been placed on the vehicle's roof. After checking with Juarez's brother and a friend, Jamie called the police. 

Two or three days before Juarez was abducted, Jeffery Logan, an escapee from a California honor farm, arrived in Tucson and met with Petitioner. On Friday, July 10, Logan went with Petitioner to a pawn shop and helped him buy a .380 semi-automatic handgun. Petitioner and Logan spent the next night driving around Tucson. At some time between 11:15 p.m. and midnight, Petitioner and Logan abducted Juarez from the parking area of his apartment complex. 

At midnight, Petitioner used Juarez's ATM card and withdrew approximately $200. He then drove Juarez out to the Silverbell Mine area near Marana. Petitioner walked Juarez fifty to seventy feet from the truck, forced him to disrobe, and shot him in the head twice. Petitioner then made two unsuccessful attempts to use Juarez's ATM card in Tucson. The machine kept the card after the second attempt. Petitioner and Logan repainted Petitioner's truck, drove to Flagstaff, and sold it. They bought another .380 semi-automatic handgun with the proceeds. 

While in Flagstaff, Petitioner and Logan met a man and woman who were traveling from California to Kansas. They abducted the couple and made them drive to Durango, Colorado; in a motel room there, Petitioner forced the man to disrobe and sexually assaulted him. 

Later, Petitioner, Logan, and the couple drove to Denver, where the couple escaped. Logan and Petitioner separated. Logan subsequently contacted the Tucson police about the murder of Juarez. He was arrested in Denver. 

With Logan's help, the police located Juarez's body. Later that day, the police arrested Petitioner at a homeless shelter in Tucson. He was carrying the handgun purchased in Flagstaff and a pair of handcuffs. After having been read his Miranda rights, Petitioner answered some questions before asking for a lawyer. He admitted that he purchased a handgun with Logan on July 10. He said that on the day of the abduction and homicide he was "cruising" through apartment complexes, possibly including the Promontory Apartments. When confronted with the ATM photographs, he initially denied being the individual in the picture. After having been told that Logan was in custody and again having been shown the photographs, Petitioner said, "I guess my life is over now."

When, during sentencing, Kemp was asked if he had anything to say in mitigation of his crime, Kemp first expressed satisfaction with his defense counsel. Kemp then offered the following:

The prosecutor, in his alleged wisdom, has portrayed me as being a killer without remorse or regret. This is a wholly inaccurate assessment. I feel a deep and abiding sense of remorse at having permitted friendship to stay my hand in the face of wiser counsel; thus electing not to kill Jeff Logan at a time when both instinct and circumstances demanded his death. 

You can rest assured that is a lapse of judgment I will never repeat and one which I will bend all my energies towards correcting in the not too distant future. Beyond that, I regret nothing. 

The media has engaged in an orgy of speculation and innuendo concerning the events of mid-July '92. They printed and reported every word spewed from Logan's mouth as though they were engraved in stone and handed down from God. They printed every accusation Logan made, whether or not it had the slightest bearing on this case, and at no time made any effort at verification. I was convicted in the press and on the televised news long before my case ever came to trial. Make no mistake, the day will come when I return to Tucson. And on that day I will remember all the kind things certain reporters had to say about me. 

The so-called victim was not an American citizen and, therefore, was beneath my contempt. Wetbacks are hardly an endangered species in this state. If more of them wound up dead, the rest of them would soon learn to stay in Mexico, where they belong. 

I don't show any mercy and I am certainly not here to plead for mercy. I spit on the law and all those who serve it; most especially you, Peasley. I have more respect for Salgado than I have for you.

I oppose the execution of people who might be factually innocent of the crime for which they are to die. In all other instances, I stand mute. In the case of Thomas Arnold Kemp, I stand mute.

The Impending Execution of Beunka Adams

Beunka Adams sits on death row awaiting execution by the people of Texas. His execution is scheduled for 26 April. I offer a summary of the case from the adverse appellate decision Adams v. Thaler (2010):

On September 2, 2002, Candace Driver and Nikki Dement were working at BDJ's convenience store in Rusk, Texas. Kenneth Vandever, a customer described as mentally challenged who often "hung around" at BDJ's and helped take out the trash, was in the store with Candace and Nikki when two masked men entered the store. One of the men was armed with a shotgun and demanded money. The two men were later identified as Beunka Adams and his co-defendant, Richard Cobb. 

After taking the money from the cash register, Adams demanded the keys to a Cadillac parked outside. After Candace produced her car keys, Adams forced her, along with Nikki and Kenneth, into the car. As Adams drove Candace's car, Nikki said, "I know you, don't I?" Adams said "Yes," and took his mask off. When they arrived at a remote pea patch near Alto, Cobb pointed the shotgun at Candace and Kenneth and Adams ordered them to get into the trunk of the Cadillac. Adams then took Nikki to a more secluded spot, away from the car, and sexually assaulted her. Later, Adams led Nikki back to the Cadillac and let Candace and Kenneth out of the trunk, but he tied the two women's arms behind their backs and made them kneel on the ground while the two robbers made their escape. Adams and Cobb seemingly developed a plan to leave Kenneth untied so that he could free the women once Adams and Cobb were far enough away from the scene. Adams, however, believed that Kenneth was attempting to untie the women too soon, so he returned and ordered Kenneth to kneel behind the women. Candace heard Kenneth say that "it was time for him to take his medicine and that he was ready to go home." 

The women then heard a single gunshot. Adams asked, "did we get anybody?" And Candace said, "No." Shortly thereafter, a second shot was fired, and Kenneth cried out, "They shot me." Kenneth Vandever died from the gunshot wound. Seconds later, Candace heard another shot, and Nikki fell forward. Candace fell forward as well, pretending to be hit. Adams approached Candace and asked her if she was bleeding. He was carrying the shotgun. Candace did not immediately answer in the hope that Adams would believe she had been killed. Adams then said, "Are you bleeding?" You better answer me. I'll shoot you in the face if you don't answer me." When Candace said, "No, no, I'm not bleeding," Adams shot her in the face, hitting her lip. 

Adams and Cobb then turned to Nikki, asking her the same questions. Adams kicked Nikki for about a minute, joined by Cobb. Then they picked her up by her hair and held a lighter to her face to see if she was still alive. Candace feigned death for fear of being shot again. She heard Cobb say about Nikki, "She's dead. Let's go." That was the only time that Candace ever heard Cobb speak. After Adams and Cobb left, Candace got up and ran barefooted down the deserted country road and banged on the door of the first house she saw.

Adams' supporters claim he did not get a fair trial. They also note that he is remorseful for his participation in the robbery that led to the murders. It seems as if Adams claimed he did not actually commit the murders, but concedes he was involved in the robbery.

I oppose those executions in which the person to be executed has some reasonable possibility of being factually innocent of the crime for which he is scheduled to die. In all other cases, I stand mute. In the case of Beunka Adams, I have no reason to doubt that he participated in the armed robbery that led to the murders. I therefore stand mute.