Wednesday, December 21, 2005

Mav's Armstrong Booed For Skins Support

Gotta love this if you're a Skins fan: Dallas Mavericks guard Darrell Armstrong was met with boos by the home crowd after shouting "How 'bout them Redskins?" before his team's game against the Timberwolves Sunday night. Armstrong, a North Carolina native and fervent Redskins fan, was also fined $1,000 by Mavericks managment.

Sunday, December 18, 2005

How Sweep It Is!

Last week the Redskins made one-third of their fumble recoveries for the season to edge Arizona. This week they made one-fourth of their interceptions and sacks for the season to crush Dallas and sweep the season series. Perhaps it's too little too late, but sweeping Dallas is a sweet salve to sooth the sting of a statistically stolen season if the Skins miss the playoffs!

UPDATE: With Atlanta's loss to Chicago tonight the Skins have moved back into the the last NFC wildcard playoff spot. Dallas, Minnesota, and Atlanta are Washington's main competition in the wildcard race and each of those teams has at least one very tough opponent remaining (Dallas vs. Carolina; Minnesota vs Chicago; and Atlanta vs Tampa Bay and Chicago). Without calculating the tiebreakers it seems the Skins have an excellent chance of making the playoffs even if they lose against the Giants or Eagles and finish 9-7.

Current Washington Post Redskins stories: Skins Mania; Wilbon; Kornheiser; Boswell.

Saturday, December 17, 2005

Satire of Life

Sometimes life seems like an extended satire.

I traveled to Roseville, Minnesota for work last week, and the "high-speed" Internet access at the Residence Inn there routinely took more than three minutes to load articles from major newspapers around the country. (Thus there was little time for blogging.)

While catching up on the mail this weekend I discovered that DHL has "lost" one of the data tapes my morgage holder routinely sends to Experian (the credit agency), so I should be extra careful about monitoring my accounts for signs of identity theft.

I bit futher in I found a letter from my employer requiring me to sign an updated benefits form--because a dollar amount on the old form was incorrectly rounded by 1 / 250 trillionths of a cent. You read that right: Someone bound by bureaucratic rules and lacking the authority to supercede them spent hundreds of dollars to print and mail corrected forms to fix a mistake that cumulatively amounted to less than a penny.

That last reminds me of a former employer who sent, over the course of a year, at least a dozen letters requesting that I mail them a self-addressed, stamped envelope to receive my last paycheck for seventy-some odd cents.

Thursday, December 15, 2005

Skins Fumble Beat

I haven't written about the Redskins or their fumble recovery problems for several weeks, but Football Outsiders has some interesting commentary in response to a reader question and in this week's power rankings.

In the first article FO shows that Greg Williams's defenses don't historically have problems recovering fumbles and also provides a breakdown of the typical offense/defense recovery rate for different types of fumbles. Offense has a big advantage recovering fumbled snaps (3/4); defense has a slight advantage recovering fumbles on runs or receptions (normally 5/8 but the defensive advantage grows when the fumble is far downfield); and sack-fumble recoveries are evenly split. Special teams fumbles aren't mentioned.

The Skins played a horrible game against Arizona and won anyway, in part because they won the fumble recovery battle decisively for the first time this season. Washington fumbled three times, recovering each one, and also recovered four of Arizona's five fumbles. On a day when Mark Brunell threw one-third of the team's total interceptions for the season, Washington was fortunate to also make one-third of its total fumble recoveries for the season. (To be fair to Brunell most of the int's were not his fault.)

Sunday's game demonstrated again the decisive role of luck in games between nearly matched teams--and why sports columnists are usually off their rockers when they call for a coach's head after his team loses a few close games or loses "the big game" a couple of times. Frankly, I thought some of the things written by Washington Post writers after the Skins' loss to San Diego (their third in a row) three weeks ago were ridiculous.

A glance at the standings shows that Washington has had a darn tough schedule this year, and that, in combination with poor fumble luck, will likely cost them a playoff spot. Clearly the team has serious flaws--such as poor depth at defensive line and wide receiver. But they've also made more progress this year than credited by the press and fans. And the team is still fixing problems from the pre-Gibbs, Snyder era.

One of these problems was Lavar Arrington's notion that his whirlybird rides with Snyder made him the "face of the team" and exempt from team rules. Another was granting Laveranues Coles the freedom to exchange his statistical "penury" with Washington (2nd most passes in the league, 6th most catches) for "riches" with the New York Jets (17th most passes, 21st most catches, same 10.6 yard average). Coles's move up cost Washington about nine million dollars in salary cap penalties--certainly enough to shop for a star defensive lineman and a solid receiver.

I've been working on a way to quantify how many wins a tough schedule costs the average team, and I hope to have my initial findings polished up and ready to post in the next few days. My model shows that the Skins have made decent--but not great--progress this year. However, if you grant that poor fumble luck cost Washington just one win they become one of the most improved teams in the league this year. One more tidbit: If last year's Skins played this year's schedule they would likely have between four and five wins at this point, instead of the seven wins they do have.

Thursday, December 01, 2005

Deadly Innocence

Jonathan Treadaway was very, very lucky.

I'm sure he didn't feel lucky on September 11, 1974 when first questioned by Phoenix police about the sodomy and suffocation of six-year-old Brett Jordan. Or on September 12 when he was arrested and charged with the murder. And when a jury sentenced him to death for the crime early in 1975 he must have thought his luck had run out. But it hadn't. Because in 1978, after three long years on death row, a second jury cleared Treadaway of all charges.

Treadaway is one of more than 100 death row "innocents" profiled by the Death Penalty Information Center, a group opposing capital punishment. Scan the DPIC List and it's quickly apparent how lucky Treadaway was to spend only three years on death row: Wilbert Lee and Freddie Pitts waited 12 years for their freedom; Jerry Banks, 5 years; Johnny Ross, 6 years; Lawyer Johnson, 11 years; Ray Krone, 10 years; Juan Melendez, 18 years; Peter Limone, 33 years. Most importantly, Treadaway escaped execution--unlike the 23 innocents scholars Hugo Bedau and Michael Radelet say were executed between 1900 and 1986.

But Treadaway's good fortune goes far beyond his relatively brief term on death row. You'll understand why shortly.

Changing Public Opinion on Capital Punishment

Death penalty abolitionists thought they had finally defeated capital punishment in 1972 when the Supreme Court ruled all existing death penalty statutes unconstitutional. And they had good reasons for optimism. Great Britain had banned executions for good in 1969. And for the first time public opinion in the United States had turned strongly in their favor. There were no executions in the United States for five years before imposition of the judicial moratorium. A 1965 Harris poll showed just 38 percent of respondents supporting the death penalty while 47 percent opposed it.

The victory was short-lived. In 1976 the Supreme Court approved new death penalty statutes that rectified the problematic aspects of the stricken ones. The dilatory death row conveyor finally jolted back into action in 1977 and delivered double-murderer Gary Gilmore to the hereafter. After a trickle of less than three executions per year through 1982, the numbers climbed steadily to nearly 100 per year in 1999 before tailing off a bit in recent years. The number of death row inmates swelled much more quickly from a low of 134 in 1973 to a high of over 3,600 in 2000 before also leveling off.

Prospects for change continued to look grim in 1992 when then-governor Bill Clinton made a very public return to Arkansas for the execution of Ricky Ray Rector. And they worsened when President Clinton was succeeded by a governor from Texas--reviled by abolitionists as the execution capital of the Western world. Public support for capital punishment peaked in 1997 when the Harris poll showed 75 percent of Americans in favor and just 22 percent opposed.

But as public support for capital punishment swelled, abolitionists were quietly preparing a new challenge based on the contention that dozens of innocents are sentenced to death row and some executed each year. The original innocence list was compiled in 1993 in response to a request by the House of Representatives Subcommittee on Civil and Constitutional Rights and built on the work of Bedau and Radelet, which was first published in the Stanford Law Review in 1987 ("Miscarriages of Justice in Potentially Capital Cases"). Politicians and the public paid increasing attention to this argument as the DPIC List was publicized, and by 2000 the Harris poll showed support for capital punishment had slipped to just 64 percent. More importantly, the average poll respondent believed that a stunning 13 percent of all convicted murderers are innocent.

Since 2000 abolitionists have racked up an impressive string of victories with the innocence tactic. In 2003, outgoing Illinois governor George H. Ryan announced that his state's capital system was broken and commuted the sentences of all 167 death row inmates. Media outlets from NPR to ABC's Extreme Makeover have showered favorable publicity on DPIC. And death row inmates found a sympathetic new spokesman in Krone, an Air Force veteran with no criminal record who was twice convicted and left languishing in prison for 10 years before being exonerated by DNA evidence in 2002. (Krone was featured on Extreme Makeover earlier this year.) Finally, in 2002 the Supreme Court struck down the death penalty for mentally retarded criminals (Atkins v. Virginia) and in March of this year banned the death penalty for juveniles (reversing its 1988 Thompson v. Oklahoma decision).

But despite these victories, the latest Harris death penalty poll, published in early 2004, found public support for the death penalty up slightly, to 69 percent. And confidence in the judicial system's fairness had also strengthened: The average respondent now believed 11 percent of all convicted murderers to be innocent--versus 13 percent in the 2000 poll.

Investigating the Innocence List

I've personally tended to ignore the capital punishment debate, considering it a well-settled issue morally, practically, and politically. By intentionally killing another person you forfeit your right to live, end of story. A large majority of Americans have agreed with this position for most of our nation's history, so why waste energy debating capital punishment when our polity is narrowly divided on many other important issues?

By now you're probably wondering why I bothered writing a lengthy article about a subject I'm ostensibly disinterested in. I've been familiar with the DPIC List for several years. (The history of the DPIC List is chronicled in depth here by Ward Campbell, a Deputy Attorney General for the State of California, who represents the state in death penalty appeals.) I was also aware of strong critiques of the List by Campbell and others, including Ramesh Ponnuru of National Review. So when a liberal relative asked me in the aftermath of the Terry Schaivo debacle how someone who is pro-life could support the death penalty, especially when so many innocent people have been executed, I decided to dig into the DPIC List and see if my previous dismissal of the List was justified.

With that in mind, I'll focus here on the assertion that we should end capital punishment because innocents are being executed. I'll only deal peripherally with other moral and pragmatic questions.

The State and Deadly Force

The first mistake many people make when thinking about capital punishment is separating it from other uses of deadly force by the state.

At its essence the modern state is, in the words of political economist Max Weber, "a human community that (successfully) claims the monopoly of the legitimate use of physical force within a given territory". Political philosophers as ideologically and chronologically diverse as Thomas Hobbes and Leon Trotsky have, despite their many other differences, at least agreed on this central definition of statehood, and also that the only realistic alternative to a state monopoly on force is anarchy (the Hobbesian kind, not the stateless cooperative existence dreamed of by modern anarchists).

By logical extension the state's use of force must include deadly force against both internal and external antagonists: If I refuse to pay taxes, I must ultimately go to jail or others will imitate me and the state may disintegrate. If I resist apprehension and offer deadly force against the agents of the state, those agents must be prepared to respond with deadly force. I don't think most people will find this reasoning controversial so I'm not going to belabor the point any further.

The state then has three primary targets against which it may use deadly force:

  • Against other states or non-state external actors (such as foreign terrorists) to preserve the security of its residents or agents (acts of war)
  • Against residents who actively resist enforcement of the laws of the state or who are in the process of threatening the security of its residents or agents (law enforcement)
  • As after-the-fact punishment for violation of the laws of the state (capital punishment)

I've slogged through the previous bit to make this point: Capital punishment, is, in America, without question the safest, most measured, cautious, just, and rarely used type of deadly force in world history. This must be considered (but usually isn't) by those who oppose executions on both general and specific moral grounds. The first type of abolitionist includes those who claim that the New Testament of the Bible implicitly rejects capital punishment, while ignoring the New Testament's clear sanctioning of states and thus by extension the deadly force necessary to sustain them. The second type includes those who reject capital punishment for its real or perceived risk to falsely convicted innocents, despite the fact that capital punishment poses less risk to innocents than any other type of deadly force exercised by the state. I'll review the relative risks posed by each type of deadly force in the next section.

Deadly Force Gone Wrong

We sometimes forget the magnitude of the collateral costs we pay for hanging on to civilization. Consider the costs of war. The "innocent" who pay are not just accidentally killed civilians. The soldiers who fight are also legally and morally innocent. (For this part of my argument let's keep things simple and assume these soldiers are not Nazi storm troopers--they're fighting a just war. Ok?)

From the 130 soldiers killed in the Moro Rebellion of 1902-1913 to the 17 sailors killed on the USS Cole in 2000 the United States suffered about 640,000 war deaths in the 20th century--well over 6,000 per year. Throw out World War I, World War II, Korea, and Vietnam and the tally still runs to more than two-dozen combat deaths per year. Toss in over a million enemy civilians killed by American forces (mostly in Germany and Japan during World War II) and the count of dead innocents gets much worse.

I realize some people consider all wars immoral. Let's step past that for now. I'm simply pointing out the problem with arbitrarily separating capital punishment from other uses of deadly force that utterly eclipse it in terms of collateral dead, even if you assume dozens of innocents are executed each year.

So what about law enforcement? It's not quite as chaotic as war, but police still kill hundreds of people each year during high-speed pursuits and in various shooting incidents. We hear about a few of the most controversial police shootings each year, such as those of Amadou Diallo in 1999 or 13-year old Devin Brown earlier this year, but police actually shoot and kill almost 400 people every year. Precise national statistics on police shootings are not available so we don't know how many of these shootings are considered to be justified. But we can look at individual cities where statistics have been examined and collected in more detail.

Just in Washington, D.C., from 1993-1998 police shot and killed 57 people and fired their weapons a total of 640 times, including 54 times at moving cars. (This last is considered especially risky because of the dangers of ricocheting bullets or an out of control vehicle to bystanders.) Ninety percent of these shootings were deemed justified after departmental review, but according to the Washington Post many of those initially justified shootings were later disputed and led to lawsuits and large financial settlements by the District.

In addition to shootings, police nationwide typically kill 300-400 more people each year in high-speed chases. More than a third of those killed were not even involved in the chase.

One might object to equating capital punishment with other types of deadly force on the grounds that capital punishment is entirely optional, while deadly force exercised during police and military actions is only used in response to proximate and deadly threats. But in fact intricate rules and regulations govern all uses of deadly force.

Policemen and soldiers must adhere to complex rules of engagement and pursuit: If I shoot a fleeing bank robber will I accidentally hit the pregnant woman across the street? How great a threat to others is the robber if he escapes? Is he known to be violent? Has he killed anyone during the robbery? If I'm threatened by an attacker with a knife how close must he be before I can shoot? When should I shoot to kill? Should I fire a warning shot first? Will this shot endanger his hostage? If I pursue this murder suspect through a residential neighborhood will I hit a child with my cruiser? If I bombard this terrorist safehouse will innocent neighbors will die? Is the enemy soldier playing dead? Is that man a civilian or an out-of-uniform enemy?

Such decisions are made rapidly, under tremendous stress, and with limited or conflicting information. The risk of mistakes here is on a whole different scale than the risk of executing an innocent in a capital crimes trial, where decisions are made over months and years, and literally millions of dollars are spent to allow every possible consideration and reconsideration of guilt and proper sentencing.

The Rare Death Penalty

In the United States today we reserve capital punishment for a tiny percentage of the most evil crimes. Before ginning up angst over supposedly unfair applications of capital punishment today, a perspective-inducing stroll through history is in order.

Sixteenth century England, the birthplace of modern American jurisprudence, offered an imaginative range of execution styles depending on the crime committed. Available methods included boiling, burning at the stake, hanging, beheading, and drawing and quartering. Capital punishment could be imposed for marrying a Jew, not confessing a crime, treason and a host of other offenses.

By the 18th century 222 crimes were punishable by death--including stealing, cutting down a tree (presumably this had more to do with overzealous protection of monarchical property than nascent environmentalism!), and poaching rabbits. In the early 1800s the English finally whittled their unwieldy list of death-worthy offenses down to about 100.

In the American colonies, early capital crimes included stealing grapes, killing chickens, trading with Indians, and striking one's parents.

Today, while much of the Western world has eliminated or severely restricted use of the death penalty, execution is still imposed for a variety of crimes in oppressive countries such as China and Vietnam. China performs over 90 percent of the world's executions each year, often for offenses such as tax evasion and corruption. Vietnam updated its list of capital offenses in 1998 to include computer hacking and environmental pollution while removing embezzlement and graft.

In countries under totalitarian rule, such as North Korea and Saddam-era Iraq, the death penalty can be arbitrarily imposed on the whim of the head of state or even minor government apparatchiks.

Understanding the history of capital punishment is important for a sense of proportion. In America we reserve execution for a small number of the most heinous crimes. In a world full of injustice and evil it's important to remember what true injustice looks like lest we waste our energies fighting what isn't injustice at all.

Are Innocents Being Executed?

Now we come to the important question: Are innocents being executed? Death penalty abolitionists say they are and point to the DPIC List as proof. But this is simply faulty logic (or polemic legerdemain): No one on the DPIC List was executed so they can't be counted as evidence of executed innocents.

The abolitionist counter-argument goes like this: We know that over 100 people on death row since 1973 have been found innocent and released. This amounts to one freed death row inmate per five executions--an error rate approaching 20 percent. Thus we can be almost certain that many more innocents have slipped through the appeals process and have been executed.

But this is yet another example of faulty logic. As Campbell points out, the DPIC List must be compared against the number of people sentenced to capital punishment, not against the number of death row inmates executed. The DPIC List includes 119 defendants sentenced to death between 1973 and 2000 while 6,930 death sentences were imposed during the same period--an error rate of just 1.7 percent. (And as we'll see in the next section the actual innocence of many of those on the DPIC List is highly disputable.) As pro-death penalty activist Dudley Sharp points out, because of the massive effort spent reviewing capital cases a convicted murderer who is actually innocent is more likely to be set free if he is sentenced to death than if he is merely sentenced to life in prison:

34% of all death penalty cases are overturned on appeal [meaning new trials are granted, most of which result in re-convictions]. Those sentenced to death are 5 times more likely to get off death row by means other than execution (appeals, clemency, other death). Even with the recent changes in habeas corpus procedure, at both state and federal levels (AEDPA), death row inmates still have 16 levels of post conviction review available. The average time on death row prior to execution is about 9 years. Because of the extreme backlog of longer serving death row inmates, it is likely that, for quite some time, such average will get longer, not shorter. For the past few years the average has been about 11 years.

If your objection to execution is the irreversible error of an innocent dying, please consider: No one disputes that death penalty cases have the greatest level of due process, what is, in fact, known as "super due process". Therefore, it is much more likely that an innocent sentenced to a life term is more likely to die, as in innocent in prison, than it is that an innocent is likely to be executed. Both irreversible error, but one much more likely than the other.

Earlier I mentioned a Bedau/Radelet study that claimed that 23 innocents were executed between 1900 and 1986. This study is frequently cited by death penalty opponents as confirming the reality of wrongful executions, but it is much less relevant to the current discussion than the DPIC List. All but one of the executions discussed in the Bedau/Radelet study occurred prior to the 1976 capital punishment reforms. And the Bedau/Radelet study has met far harsher criticism than even the DPIC Innocence List, primarily because of sloppy methodology that based innocence determinations on defendant or defense attorney claims of innocence, newspaper articles, or prosecutorial misconduct. Justice Department lawyers Stephen Markman and Paul Cassell reviewed 13 of the 23 cases (only the post-1950 cases for which original court documents were available) in 1988 and concluded that

Bedau and Radelet's catalogue of "innocents" ignores physical evidence of guilt, incorrectly cites sources that in fact indicated defendants were guilty, includes works of fiction as proving innocence, and contains other serious flaws.

(Markman has served on the Michigan Supreme Court since 1999, while Cassell is a professor at the University of Utah College of Law.) Suffice it to say that when challenged on their work, Bedau and Radelet responded, "We agree with our critics that we have not proved these executed defendants to be innocent; we never claimed that we had." (41, 1 Stanford Law Review, 11/1988).

Problems with the DPIC List

The DPIC List is most commonly criticized for intentionally confusing de facto and de jure innocence. Campbell's case-by-case critique shows that 68 of the 102 defendants on the list as of 2000 are either not definitively innocent or were included on the list despite being convicted before the 1976 reforms. He argues that there is strong evidence of actual guilt in many of the DPIC cases. In many of these cases the original conviction was overturned on appeal, and the defendant was not re-convicted only because evidence presented at the original trial was disallowed at subsequent trials; or prosecutors elected not to retry the case; or key witnesses were no longer available to testify. Let's review some of these cases.

Defendant Clarence Smith was listed on the DPIC List because he had been acquitted in Louisiana state court in 1995--despite having later been convicted on the same charges in federal court. When Smith's federal conviction was pointed out to DPIC, his name was replaced with a brief explanation, but the explanation has since been removed. (Note that there is another Clarence Smith on the List who was convicted in New Mexico and freed in 1976.)

Another defendant, Larry Osborne, was convicted of breaking into the house of an elderly couple, beating them to death, and burning their house down. A friend and possible accomplice testified against Osborne in the grand jury proceeding that preceded his trial; however, the friend drowned before the actual trial. The dead witness's taped testimony was played for the jury and Osborne was convicted, but his conviction was overturned on appeal because the defense was unable to cross-examine the dead witness. Osborne was then acquitted at his retrial where the taped testimony was disallowed.

In a similar case, Jeremy Sheets was convicted of raping and murdering a young black girl, partly on the basis of secretly tape-recorded statements made by his friend and accomplice, Adam Barnett. The recordings implicated both Sheets and Barnett. The two were arrested after the recordings were turned over to police by a third person. Barnett then committed suicide while in prison awaiting the start of the trial. Sheets was convicted, but just as in the Osborne case his conviction was overturned on appeal because the defense could not cross-examine the dead Barnett whose statements implicated Sheets. The prosecutor doubted he could prove Sheets's guilt beyond a reasonable doubt without the taped testimony and decided not to retry Sheets, though there was other evidence of his guilt including his own contradictory testimony.

Perhaps the biggest problem with the DPIC List is that it often significantly mischaracterizes questionable cases to make the defendants appear less guilty. This is quickly evident when you compare the DPIC descriptions of the cases mentioned above with Campbell's more complete descriptions. But who is being more honest? To find out I decided to check other source material for a couple of the DPIC cases and compare the DPIC description of each case with Campbell's and Ponnuru's descriptions. And that brings us back to Jonathan Treadaway.

Lucky Jonathan Treadaway

Here's how DPIC describes the Treadaway case:

Jonathan Treadaway Arizona Conviction 1975 Acquitted 1978
Treadaway was convicted of sodomy and first degree murder of a six-year-old and sentenced to death. The conviction was overturned, and he was acquitted of all charges at retrial by the jury after 5 pathologists testified that the victim probably died of natural causes and that there was no evidence of sodomy. Members of the jury reported noted that prosecutors had failed to prove that Treadaway was even inside the victims' home. (State v. Treadaway, 568 P.2d 1061 (1977))

Read this and it sounds like Treadaway barely escaped a terrible injustice. But this description raises more questions than it answers: What made the police suspect Treadaway in the first place? Who were the five pathologists who testified on Treadaway's behalf? What did the first jury find so convincing that so failed to impress the second jury? Campbell's article answers many of these questions and presents the case quite differently:

Jonathan Treadaway-- State v. Treadaway, 568 P.2d 1061, 1063-1065 (Ariz. 1977); State v. Corcoran (Treadaway I) 583 P.2d 229 (Ariz. 1978) (Treadaway II). Treadaway was convicted of the sodomy and first degree murder of a young boy in the victim's bedroom. His conviction was reversed and he was acquitted on retrial.

Treadaway's two palmprints were found outside a locked bedroom window of the victim's home. When Treadaway was arrested, he had no explanation for these palmprints. Treadaway admitted being a peeping tom in the victim's neighborhood, but did not remember ever looking in the victim's house. He denied being at the victim's house the night of the murder. However, the victim's mother testified she washed the windows the day before the murder, "raising an inference that the palm prints found on the morning after the murder [were] fresh" and also raising the inference that Treadaway was lying. Pubic hairs on the victim's body were similar to Treadaway's. His conviction was reversed by the Arizona Supreme Court in a 3-2 decision because the trial court erroneously admitted evidence that Treadaway committed sex acts with a 13-year old boy three years before the murder.

When Treadaway's retrial began, the Arizona Supreme Court reviewed several pretrial evidentiary rulings. It admitted evidence that Treadaway sexually attacked and tried to strangle a boy three months before the murder at issue in the boy's bedroom. However, the court excluded the interrogation in which Treadaway failed to explain his palmprints outside the victim's bedroom window, specifically refused to provide information any information, and made other incriminating statements. The exclusion was based on the police failure to comply with the technical requirements of the Miranda decision, not because Treadaway's statements or failure to explain the palmprints on the window were somehow unreliable or involuntary.

This decision to exclude Treadaway's interrogation was a crucial difference between his two trials. Although there was defense evidence that the victim died of natural causes, the jurors who acquitted Treadaway on retrial later stated that they were actually concerned about the lack of evidence that Treadaway had been inside the boy's home. Stanford, at 164; In Spite of Innocence, at 349. Therefore, Treadaway's failure to explain the palmprints at the window could have been critical evidence since those palmprints at the very least would have connected Treadaway with a location just outside the boy's home on the night of the murder. Treadaway's inability to explain the suspicious presence to the police of his fingerprints would ordinarily indicate a "consciousness of guilt" about his presence at the boy's home. However, the jury was never permitted to know that Treadaway had had no explanation for those palmprints-a circumstance consistent with his guilt. Thus, significant probative evidence of Treadaway's consciousness of guilt about the palmprints on the windowsill, directly relevant to the jury's concern about the case, was never disclosed to the jury at his second trial. Since it cannot be known what the impact of that excluded evidence would have been on the second jury, Treadaway's acquittal on retrial did not demonstrate that he was innocent.

Furthermore, in light of the recent United States Supreme Court decision in Ring v. Arizona it is speculation whether a jury would have found Treadaway eligible to be sentenced to death.

This sounds much worse for Treadaway. It certainly puts his case in the same category as Smith, Osborne, and Sheets--who all, but for fortuitous twists of circumstance, seemed more likely to be found guilty than innocent. We may never know the whole truth about these cases. But we do know that many of the DPIC "innocents" look a lot less innocent the more you know about the circumstances behind their acquittals.

Sometimes that's all we'll ever know. But not in this case. Because when I started searching for source documents on the Treadaway case I found something quite astonishing: A message posted to this death penalty discussion forum late last year by Cathy Jordan, the mother of the six-year-old boy allegedly sodomized and murdered by Jonathan Treadaway in 1974.

Jordan claimed Treadaway's case had been reopened by a Phoenix cold-case detective named Ed Reynolds, and that Reynolds was closing in on Treadaway with DNA evidence of his guilt when Treadaway died of a terminal illness. According to Jordan, the new evidence implicated Treadaway in the killing of her son Brett and a second child, a three-year-old girl named Doris Morris.

I tried but was unable to reach Detective Reynolds at the Phoenix police department to confirm the story. However, I was able to obtain department records of the relevant investigations through a public records request. (I've scanned the key documents and archived them here.) The records show that Reynolds first reopened the 1974 Morris case in June, 1994.

Reynolds spent two years examining documents and reviewing evidence in the 20-year-old case, and in 1996 put Treadaway under surveillance to obtain a DNA sample for testing. In June, 1996 preliminary DNA tests performed on a small, mixed tissue sample from the Morris murder scene matched DNA taken surreptitiously from Treadaway.

For unknown reasons Reynolds's work on the case slowed again, though he continued to search evidence archives for more tissue samples and request additional tests from the FBI's crime lab. Finally, in October, 2000 Maricopa County Superior Court Judge Susan Bolton issued a warrant authorizing Reynolds to apprehend Treadaway and take blood samples for confirmation of the previous DNA tests.

When Reynolds tried to arrest Treadaway at his residence he discovered that the 47-year-old had recently died of liver failure. Based on the DNA test results and other evidence implicating Treadaway, Reynolds closed the Morris case. A Phoenix-area journalist who has written several stories on cold-case crimes solved by Reynolds confirmed to me that Reynolds was certain his new evidence was enough to convict Treadaway in the Morris case. Having once been found innocent, Treadaway could not have been retried for the Jordan murder.

In a couple of strange twists, it turns out that Reynolds's niece married one of Brett Jordan's now-adult, younger brothers. And according to Cathy Jordan, Reynolds himself opposes the death penalty.

But beyond the refutation of Treadaway's innocence, the real story here is the impact of the DPIC List on murder victims' families in cases where the accused murderer seems to have escaped punishment by sheer luck. Stricken with the pain of missing loved ones, they now must watch as DPIC transforms the probable killers into aggrieved victims. In the words of Cathy Jordan:

It never quits hurting. A song, a smell, a little boy in a Boy Scout troop. One of my sons keeps a loaded gun under one of his pillows.

My husband and I had woke up, and he left for work, and I walked back to my bedroom. Passing Brett's room, I saw him facing away from me. His eyes were closed. He was laying on his stomach. I touched him, and he was so very cold. That day changed everyone in my family's lives. Shortened some. You can never feel happy all the way because something is missing. I liken it to losing a leg. I learned to live with it, but I'm aware of it all the time. If someone told me I could hold his sweet precious body again long enough to give up a limb--if I could just hold him and bleed to death I'd do it.

To have [Treadaway] as a poster boy against the death penalty nauseates me.

I informed Richard Dieter, Executive Director of DPIC, of the new information in Treadaway's case and asked him if he would remove Treadaway from the List if I provided him with results from the DNA test that implicated Treadaway in the Morris murder. He responded that DPIC would only remove Treadaway from the list if his guilt were conclusively proved for the original crime for which he was convicted and released.

When asked what would constitute such "proof", Dieter said he would only accept, "independent verification of the DNA results inculpating Treadaway in the murder, with the conclusions published in a public record such as an article in a reputable newspaper or court proceeding". Of course, this is not even as charitable as it sounds at first reading. Treadaway is dead so he is hardly likely to be tried for these crimes. Independently duplicating the DNA results would presumably require a new tissue sample from Treadaway--again not likely as this would require his exhumation and would be impossible if he were cremated. Also, there might not be enough remaining tissue from the crime scene to perform a second DNA test.

Finally, even if Treadaway were alive he could not be tried again for the Jordan murder because of the double jeopardy clause of the Fifth Amendment. So bizarrely, under DPIC's standards, Treadaway could be implicated in the Morris murder by DNA evidence, and, if he were still alive, tried and convicted of the Morris murder, and still remain on the DPIC Innocence List because he was never re-convicted of the crime that originally sent him to death row .

In some circumstances conviction of a second murder might be immaterial to the original murder, but even that's not true in the Treadaway case. Doris Morris was murdered just 11 days before Brett Jordan, in a nearby area, and with the same modus operandi. Treadaway was the prime suspect in the original Morris investigation but apparently was never charged due to a lack of physical evidence.


UPDATE: A couple of weeks after posting this article I was tipped off to a write-up published by The Arizona Republic shortly after Reynolds attempted to arrest Treadaway:

26-YEAR-OLD CASES OF CHILD MURDER SOLVED WITH DNA TEST, POLICE SAY

By Judy Villa

December 18, 2000

DNA tests have linked the same man to a pair of child murders that rocked Maryvale 26 years ago.

Doris Morris, 3, and Brett Jordan, 6, were sodomized and suffocated in their beds 11 days apart in August 1974.

The suspect, Jonathon Charles Treadaway Jr., was tried twice for Brett's murder and acquitted the second time. He died over the summer before the final DNA tests could be completed in Doris' death.

But as a result of preliminary DNA tests, there is "no doubt in my mind" that Treadaway committed both murders, Phoenix police Detective Ed Reynolds said.

The tests, conducted more than two decades after the murders, linked Treadaway to semen found in Doris' underpants and hairs found on Brett's body, he said.

Police, confident of the test results, closed the investigation of Doris' murder in October, after Treadaway died.

The case demonstrates the power of DNA technology. And it shows that even though DNA can exonerate the innocent, it also can be equally effective in tagging the guilty.

...

Reynolds said he focused on the Doris Morris murder because Treadaway couldn't be tried again for the Brett Jordan murder -- that would have been double jeopardy.

Treadaway initially was convicted -- based primarily on a palm print and a hair that microscopically matched his in color and texture -- and sent to death row for killing Brett. But the Arizona Supreme Court threw out the conviction and he was freed after a second trial in 1978.

Reynolds called the move "an extreme injustice" and said "it was obvious to me he was guilty."

"I wanted to see him back in prison again," Reynolds said. "Murder is murder, but when you murder young children, that's about as heinous as it gets."

Phoenix detectives realized in 1996 that they had evidence hairs and semen from Doris' murder that they could use to perform DNA tests. But it would be another two years before DNA testing progressed enough to link Treadaway to that evidence, Reynolds said. Even then, the sample provided only six sequences of DNA -- not quite enough to take to court.

The final test, which would have used blood instead of saliva, would have been the "gold plating" on the case, Reynolds said. But when detectives got a court order in October to draw Treadaway's blood, they learned he had died from cirrhosis of the liver.

Reynolds was disappointed, but he said Treadaway didn't get off scot-free.

"I have no doubt in my mind that he's paying for it today," Reynolds said.

The article was locked up in the Republic's paid archives and for some reason didn't show up in my LexisNexis searches. It confirms several of my speculations that I couldn't confirm with Reynolds.


Delbert Tibbs: Murderous Minister?

Next I reviewed the case of Delbert Tibbs. Here's what the DPIC List says about Tibbs:

Delbert Tibbs Florida Conviction 1974 Charges Dismissed 1977
Tibbs was sentenced to death for the rape of a sixteen-year-old white girl and the murder of her companion. Tibbs, a black theological student, was convicted by an all-white jury on the testimony of the female victim whose testimony was uncorroborated and inconsistent with her first description of her assailant. The conviction was overturned by the Florida Supreme Court because the verdict was not supported by the weight of the evidence, and the state decided not to retry the case. Tibbs' former prosecutor said that the original investigation had been tainted from the beginning and that if there was a retrial, he would appear as a witness for Tibbs. (Tibbs v. State, 337 So.2d 788 (Fla. 1976)).

This sounds like Tibbs got a pretty bad deal. According to DPIC, Tibbs' prosecutor seemed to think he was innocent! Who is this prosecutor who would seek a conviction against someone he believed to be innocent? We don't know because the DPIC site doesn't give his name. In comparison, here's Campbell's description of the case:

Delbert Tibbs--Tibbs v. State, 337 So.2d 788 (Fla. 1976) (Tibbs I); State v. Tibbs, 370 So.2d 386 (Fla.App. 1979) (Tibbs II); Tibbs v. State, 397 So.2d 1120, 1123 (Fla. 1981) (Tibbs III); Tibbs v. Florida, 457 U.S. 31 (1982) (Tibbs IV). Tibbs was convicted of raping a woman and murdering her boyfriend. The chief witness was the surviving rape victim who identified Tibbs as her boyfriend's murderer.

Tibbs' conviction was reversed by a 4-3 vote of the Florida Supreme Court. The majority applied an anachronistic review standard that "carefully scrutinized" the testimony of the prosecutrix since she was the sole witness in the rape case "so as to avoid an unmerited conviction." Tibbs I at 790. The conviction was not even reversed because the Florida court found the evidence legally insufficient, but merely because the Florida court found the "weight" of the evidence was insubstantial. The court found the prosecutrix's testimony to be doubtful when compared with the lack of evidence (other than her eyewitness testimony) that Tibbs was in the area where the rape-murder occurred. Id. at 791.

Subsequently, in a later opinion, the Florida Supreme Court repudiated this "somewhat more subjective" rule that permitted an appellate court to reverse a conviction because of the weight of the evidence, rather than its sufficiency. In hindsight, the Florida Supreme Court candidly conceded that it should not have reversed Tibbs' conviction since the evidence was legally sufficient. Tibbs III at 1126. The old review standard applied to Tibbs' original case was a throwback to the long discarded rule that a rape conviction required corroboration of the rape victim's testimony-an unenlightened rule which inherently distrusted the testimony of the rape victim. Id. at 1129 fn. 3 (Sundberg, C.J. dis. & conc.); see e.g. People v. Rincon-Pineda, 14 Cal.3d 864 (Cal. 1975). The reversal of Tibbs' conviction was a windfall for Tibbs, not a finding of innocence.

Subsequently, a debate in the Florida courts as to whether or not Tibbs could be retried under the Double Jeopardy Clause made its way to the United States Supreme Court. Justice O'Connor's opinion explained that the rape victim gave a detailed description of her assailant and his truck. Tibbs was stopped because he matched her description of the murderer. The victim had already viewed photos of several single suspects on three or four occasions and had not identified them. She examined several books of photos without identifying any suspects. However, when she saw Tibbs' photo, she did identify Tibbs as the rapist-murderer. She again identified Tibbs in a lineup and positively identified him at trial. Tibbs IV at 33 & fn. 2. At trial, the victim admitted drug use and that she used drugs "shortly" before the crimes occurred. She was confused as to the time of day that she first met Tibbs. Although not admitted as evidence, polygraphs showed however that the victim was truthful. Tibbs denied being in the area during the time of the offense and his testimony was partially corroborated. However, the prosecution introduced a card with Tibbs' signature which contradicted his testimony as to his location. Tibbs disputed that he had signed the card. Id. at 34-35. O'Connor's opinion also noted the evidence that the Florida Supreme Court had originally believed weakened the prosecution's case. However, since the evidence of guilt was not legally insufficient, the Double Jeopardy Clause did not bar Tibbs' retrial. Id. at 35.

Ultimately, due to the current status of the surviving victim--a lifelong drug addict--the original prosecutor concluded the evidence was too tainted for retrial. In Spite of Innocence, at 59. Nonetheless, the evidence recounted in the United States Supreme Court decision hardly supports a claim that Tibbs is actually innocent.

The state prosecutor who chose not to retry Tibbs recently explained to the Florida Commission on Capital Crimes that Tibbs "was never an innocent man wrongfully accused. He was a lucky human being. He was guilty, he was lucky and now he is free."

Campbell's description of the case is certainly more thorough. Note that the prosecutor's statement about tainted evidence is framed very differently in the two descriptions of the case. According to DPIC he seems to think the whole trial was a farce. But Campbell's version of the prosecutor's statement clarifies that the evidence was too tainted for retrial due to substance abuse by the key witness, and goes on to say that Tibbs was guilty and simply got lucky.

I suppose there could have been more than one prosecutor involved--neither description makes this completely clear--but again, why would the first prosecutor pursue the case if he really was willing to serve as a defense witness?

This case history assembled by the Florida Commission on Capital Cases describes the original charges, the appeal, and key issues of the case. It also includes statements from key participants in the trial and notes which participants' statements are missing and why. It indicates that only one prosecutor, State Attorney Joseph Alessandro, was involved with the case and his full statement reads:

By the time of the retrial, witness/victim Cynthia Nadeau had progressed from a marijuana smoker to a crack user and I could not put her up on the stand, so I declined to prosecute. Tibbs, in my opinion, was never an innocent man wrongfully accused. He was a lucky human being. He was guilty, he was lucky and now he is free. His 1974 conviction was not a miscarriage of justice.

The Florida Commission document also includes statements from Assistant State Attorney Dean Plattner and Raymond Marky of the Florida Attorney General's office indicating their strong beliefs that Tibbs was guilty based on their knowledge of the case, and that no one else was ever prosecuted for the crime.

I found no Treadaway-like bombshell revelations in the Tibbs case. But the DPIC description is woefully inadequate and misleading inverts the meaning of the prosecutor's statements.

A Poor Record on Innocence

Beyond inaccuracies on the current DPIC List, death penalty abolitionists have shown themselves willing marks for smooth-talking killers in some other high-profile capital cases.

In 1981, Roger Keith Coleman raped and slit the throat of his sister-in-law while her husband was at work. He was convicted and sentenced to death on the basis of evidence that included bloodstains on his pants, semen matching his rare blood type, and witness testimony which implicated him in various ways.

Coleman had previously served a 20-month prison term for attempted rape. Yet in 1990 a former ACLU lawyer named Kitty Behan believed Coleman's pleas of innocence and launched an all-out effort to free him, which included court-ordered DNA tests on the semen. When the DNA tests performed by Behan's handpicked expert implicated Coleman she hired another DNA expert to dispute the results and publicly accused a different man of the murder. (The man she accused sued for libel and reportedly won an out-of-court settlement from Behan and her law firm, according to a May 10, 2000 Wall Street Journal article.)

In 1992, shortly before he was executed, TIME and Newsweek both ran stories portraying Coleman as innocent. Neither magazine mentioned the DNA evidence confirming his guilt. Thirteen years later, dozens of anti-death penalty sites across the Web still portray Coleman as an innocent victim of a callous capital crimes system, and newspapers and anti-death penalty activists continue to pressure Virginia's current governor, Mark Warner, to allow new DNA tests. A Washington Post editorial from December of last year at least acknowledges the earlier DNA test which confirmed Coleman's guilt; but dismisses it with that vague, all-purpose pejorative of "controversial" that newspapers use when they don't have any substantive criticisms on a particular issue.


UPDATE: In January, 2006, just six weeks after I posted this article, Warner acceded to demands for a new test of the Coleman case evidence. The result: An unsurprising re-confirmation of Coleman's guilt and the gullibility of death penalty opponents who are certain a convicted murder would never lie on his way to the electric chair. The Post, not surprisingly, is certain that the death penalty is still a threat to innocents.



image
May 18, 1992 TIME magazine cover featuring Roger Keith Coleman.

American abolitionists aren't the only ones who have been duped. Great Britain permanently banned the death penalty in 1969 after the sensational trial and hanging of James Hanratty in 1962 spun up an anti-death penalty furor. Hanratty was convicted of murdering Michael Gregsten and raping and shooting Gregsten's girlfriend, Valerie Storie, after the longest murder trial in English history. Storie, shot 5 times and left for dead, survived to identify and testify against Hanratty. Discrepancies in original news accounts of Storie's testimony and Hanratty's pleas of innocence spawned a group of family members and other death-penalty opponents dedicated to proving his innocence--the "A6 Committee".

In 1967 the original suspect in the crime, Peter Alphon, confessed to the murder after reportedly being paid L25,000 by members of the Committee. Alphon later recanted his confession, but the damage was done: Parliament eliminated the death penalty in response to the continuing public outcry. The A6 Committee continued to press for official declaration of Hanratty's innocence, and in 2000 Hanratty's body was exhumed and DNA tests performed against semen obtained from Storie's underpants.

The DNA results confirmed Hanratty's guilt and cleared Alphon, the false confessor, as a suspect in the crime. Despite this, the A6 Committee continues to maintain Hanratty's innocence, claiming the DNA evidence must have been contaminated. The appellate court judgment addresses this claim at length, describing all the steps followed in handling and testing the various pieces of evidence and finally dismissing the contamination claim thusly:

With regard to the knicker fragment we have what Dr Whitaker would describe as a typical distribution of male and female DNA following an act of sexual intercourse leading, to the obvious inference that the male contribution came from James Hanratty. For that not to be the case we would have to suppose that the DNA of the rapist, also of blood group O, had either degraded so as to become undetectable or had been masked by James Hanratty's DNA during the course of a contaminating event. Moreover, we would also have to suppose that Valerie Storie's DNA had remained in its original state, or at least detectable, and had escaped being overridden by DNA from James Hanratty. The same would have to be true of the DNA attributed to Michael Gregsten. Finally, we must visualise a pattern which is wholly consistent with sexual intercourse having taken place in which Valerie Storie and James Hanratty were the participants.
The judgment concludes with this strong affirmation of Hanratty's guilt:
In our judgment for reasons we have explained the DNA evidence establishes beyond doubt that James Hanratty was the murderer. The DNA evidence made what was a strong case even stronger. Equally the strength of the evidence overall pointing to the guilt of the appellant supports our conclusion as to the DNA.
...
Just let us consider some of the more striking coincidences in the light of the DNA evidence if James Hanratty was not guilty. He was wrongly identified by three witnesses at identification parades; first as the person at the scene of the crime and secondly (by two witnesses) driving a vehicle close to where the vehicle in which the murder was committed was found; he had the same identifying manner of speech as the killer; he stayed in a room the night before the crime from which bullets that had been fired from the murder weapon were recovered; the murder weapon was recovered from a place on a bus which he regarded as a hiding place and the bus followed a route he could well have used; his DNA was found on a piece of material from Valerie Storie's knickers where it would be expected to be if the appellant was guilty; it was also found on the handkerchief found with the gun. The number of alleged coincidences means that they are not coincidences but overwhelming proof of the safety of the conviction from an evidential perspective.

An Amp of Prevention

It seems that any discussion of capital punishment must inevitably consider its role in preventing homicides. Some, such as Emory University professor Paul Rubin, claim to quantitatively demonstrate that capital punishment has a very large effect in preventing murders. Death penalty opponents, on the other hand, have been so audacious as to claim that executing murderers leads to more murders!

I'm going to take a slightly different tack and dismiss these types of econometric models out of hand--at least where they concern the relationship between executions and the murder rate. The problem is, as Rutgers sociologist Ted Goertzel discusses here, that broad econometric analyses of public policies such as capital punishment depend on too many variables to control, rely on incomplete data sets, and usually cannot be tested for their predictive ability.

For example, criminologist Anthony Harris estimates that the homicide rate would be roughly three times higher today if the quality of emergency medical care had remained at the level it was in 1957. Along the same lines, a study published by Dr. Don Soo Chon in 2002 attempted to demonstrate (with limited success) that the differences in homicide rates between many nations could be partially explained by different distributions of medical resources. I point out these two studies simply as examples of the range of variables that can come into play when studying the effects of public policy over time and across broad geographic regions.

If we agree that the death penalty is a morally appropriate form of punishment for premeditated murder--as most Americans do believe--then our choice to use it should not depend at all upon its preventative effects. However, just for the sake of argument let's embark on a short, non-econometric discussion of capital punishment's preventative qualities and weigh them against the risks of executing an innocent.

Advocates of capital punishment often mistakenly argue that capital punishment must have a large or measurable preventative effect on future murders for its preventative qualities to be considered at all. But why must this be so? We can show a high probability of some deterrent effect while acknowledging that we can never exactly measure its size.

Consider that the vast majority of criminals prefer prison to execution as evidenced by their efforts to avoid the latter. Consider also that in a large enough population any increase in the cost of a good or activity will reduce its demand or frequency. (No, execution is not a Giffen or Veblen good!)

To use a simple example, shoppers will definitely buy more packages of Oreos for 99 cents than for $1.99. And if our population of shoppers is large enough, even a one-cent cost increase is statistically guaranteed to reduce the number of packages sold. It's difficult to believe that such a tiny price increase could change anyone's mind, as a penny doesn't matter much to most of us. But even a small increase in cost will change someone's mind. Perhaps it's the ten-year-old who counts his change and finds he's just a penny short. Or maybe the old man who remembers when he could buy the same size package for a nickel, who bought a package last week for 89 cents, and who refuses, simply refuses to pay a whole dollar! That one-cent difference may matter to one person in a thousand or just one in a million, but it will matter.

We also know psychological effects can amplify the perceived variance of certain price points--which is why we see so many prices ending with nines. A price change from 99 cents to one dollar will dissuade more buyers than a change from 88 cents to 89 cents.

Death is definitely a more "costly" punishment for murder than life in prison. It's also an extremely significant psychological "price point". Is it then credible to argue that in a U.S. population of 300 million people, with about 20,000 murders per year, that not even one person will change their mind about committing a premeditated murder when faced with the prospect of execution instead of a long prison sentence?

I think we can be sure just from anecdotal evidence that far more than one murder per year is deterred by the death penalty. Consider Senator Dianne Feinstein's (D-CA) 1993 statement before the Senate Judiciary Committee:

"In the 1960's, I was appointed to one of the term-setting and paroling authorities and sat on some 5,000 cases of women who were convicted of felonies in the State of California. I remember one woman who came before me because she was convicted of robbery in the first degree, and I noticed on what is called the granny sheet that she had a weapon, but it was unloaded. I asked her the question why was the gun unloaded and she said, so I wouldn't panic, kill somebody and get the death penalty.

"That case went by and I didn't think too much of it at the time. I read a lot of books that said the death penalty was not a deterrent. Then in the 1970's, I walked into a mom-and-pop grocery store just after the proprietor, his wife and dog had been shot. People in real life don't die the way they do on television. There was brain matter on the ceiling, on the canned goods. It was a terrible, terrible scene of carnage.

"I came to remember that woman because by then California had done away with the death penalty. I came to remember the woman who said to me in the 1960's, the gun was unloaded so I wouldn't panic and kill someone, and suddenly the death penalty came to have new meaning to me as a deterrent."

(Statement of the Honorable Dianne Feinstein, Senator from California, Hearing Before the Senate Judiciary Committee on S.221 (April 1, 1993).)
Or this statement from B. Rey Shauer, former Justice of the Supreme Court of California:
That the ever present potentiality in California of the death penalty, for murder in the commission of armed robbery, each year saves the lives of scores, if not hundreds of victims of such crimes, I cannot think, can reasonably be doubted by any judge who has had substantial experience at the trial court level with the handling of such persons. I know that during my own trial court experience...included some four to five years (1930-1934) in a department of the superior court exclusively engaged in handling felony cases, I repeatedly heard from the lips of robbers...substantially the same story: 'I used a toy gun [or a simulated gun or a gun in which the firing pin or hammer had been extracted or damaged] because I didn't want my neck stretched.' (The penalty, at the time referred to, was hanging.)
I'm not arguing that criminals will rationally evaluate their chances of being caught and punished for murder. Behavioral research shows that emotion usually overrides cognitive reasoning in most decisions involving risk, even those made by rational people trying to minimize the influence of their emotions. The same research seems to indicate that the prospect of execution is just the sort of emotion-engaging cost that would deter many potential murderers.

Two elements in particular shape the emotional evaluation of risk. The first is the ability to vividly imagine a decision's outcome. This applies to both positive and negative decision outcomes and is inversely proportional to the abstractness of the risk. For example, people are willing to pay more for airline travel insurance covering death from "terrorist acts" than death from "all possible causes" simply because the former is specific and thus more readily visualized than the latter. Anecdotal presentation of risk or its connection to specific individuals also heightens tendencies to vividly imagine its possible outcomes.

The second element is an insensitivity to probability variation that is directly proportional to the vividness or extremity of the outcomes associated with a risk. Again, this applies to both positive and negative outcomes. For example, a decision to gamble or buy lottery tickets is driven by the size of the prize much more than the probability of winning. One chance in 100,000,000 of winning $1,000,000 will be seen as more desirable than 1 chance in 1,000,000 of winning $10,000, even though the average outcome in both cases is ten cents. The larger amount of money would allow a significant change in lifestyle for most people and the mental imagery of this lifestyle change overpowers the cognitive decision process. This also produces a dramatic overweighting of tiny probabilities of negative outcomes, especially when comparing them to zero probability of a vividly adverse event. For example, one study found that people were willing to pay dramatically more to reduce the chances of being poisoned by an insect spray from 5 in 10,000 to 0, than from 15 in 10,000 to 5 in 10,000 (this is known as the certainty effect ). (Risk as Feelings; Lowenstein, Weber, Hsee, and Welch; Psychological Bulletin, Vol. 127; 274-277)

This is all highly relevant to the decision process of a potential murderer and his evaluation of the risk of capital punishment. Execution is an extreme and vividly imaginable outcome, especially compared to an unknown number of years that might be spent in prison. (The shift to lethal injection and away from more gruesome means of execution has likely lessened its vividness to some unknown degree.) Anecdotal examples of this outcome are highly publicized many times each year and certainly have personal impact to many in the criminal world (i.e. the executed was a direct, once, or twice removed acquaintance). Finally the extremity of the outcome certainly leads to an overweighting of its probability in many cases. To completely remove the risk of execution by ending capital punishment would likely dampen its preventative effect much more than did its severe restriction by the Supreme Court in 1972.

One might argue that fear-inducing emotion cannot stop the decision to murder because murderers are sociopaths and thus lack the capacity to feel fear and guilt. But not all murderers are sociopaths. And I'm not trying to prove a large deterrent effect or even saying there is one in all types of cases. I'm simply showing that it's unreasonable to argue for no deterrent effect.

Especially relevant here is the Supreme Court's decision this year in Roper v. Simmons to ban the death penalty for juveniles. That case involved seventeen-year-old Christopher Simmons who, with two friends, broke into the home of Shirley Crook, bound her feet and hands with electrical wire, wrapped her head in duct tape, and threw her from a railroad trestle into the Meramac River where she drowned.

Simmons assured his accomplices that they could "get away with" the murder because they were minors. The Court's ruling eliminated all doubts on this point that may have restrained future Simmonses. Consider that juveniles aged 14-17 have typically committed between 10 and 20 percent of all murders in recent years, and we can be just as certain that Justice Kennedy signed the death sentence of more than one American when he erased the vivid specter of execution from the minds of thousands of young, potential killers.

Final Thoughts

Capital punishment abolitionists believe for various reasons that the death penalty is not a morally appropriate punishment for any crime, including the most heinous torture and slaughter conceivable. We should abandon capital punishment, they say, because it's too risky. The consequences of a mistake are irrevocable. And yet, there are and will continue to be many cases where the possibility of a mistake is zero.

Consider Brian Nichols, whose Atlanta courtroom escape and murder-spree were recorded on numerous security cameras and witnessed by dozens of bystanders. Nichols has already pleaded not guilty and will likely fight a years-long battle of appeals after his (almost certain) death sentence judgment. Or consider John Wayne Gacy, who was executed in 1994 for the murder of 33 young men. Gacy confessed to his crimes, but then gamed the system with 523 appeals over 14 years, none of which were based on claims of innocence.

Most Americans do believe that execution is a morally just penalty for premeditated murder. Opponents of capital punishment, finally realizing that their chances of swaying public opinion against the death penalty on moral grounds alone are vanishingly small, have resorted to propagandizing against the ability of the capital crimes system to deliver fair and just decisions. This tactic is abhorrent not only for its fundamental dishonesty, but for its undermining effect on the entire legal and judicial system.

Even if we accept the claims of abolitionist groups such as DPIC at face value, the risk to innocents from capital punishment is insignificant compared to the risks posed by other state exercises of deadly force, such as police and military actions. Indeed the most generous weighting of capital punishment's risk hardly balances the most conservative estimates of its preventative effects. The DPIC List represents, at worst, an erroneous conviction rate of less than two percent in capital trials. The true error rate seems closer to half of one percent (fewer than two per year) when you remove defendants who would likely have been found guilty but for procedural mistakes or good fortune. Remember, that's the wrongful conviction rate, not the wrongful execution rate. The wrongful execution rate, so far as we know, is zero. It's hard to imagine that a real-world capital crimes system could be fairer than that.

If death penalty opponents truly care to improve the fairness of the American justice system they would be wiser to drop their fixation with capital cases and turn their energies toward helping innocent defendants accused of all serious crimes.

Thanks for links from: LaShawn Barber, JunkyardBlog, Clayton Cramer, Stop The ACLU
Current Washington Post death penalty articles: 1,000th execution. No break for Tookie. Politics of punishment. More doubtful on death penalty.

Monday, November 21, 2005

Krauthammer Mangles ID

Charles Krauthammer doesn't exactly provide an honest treatment of so-called intelligent design in his most recent column. Krauthammer says ID is

a self-enclosed, tautological "theory" whose only holding is that when there are gaps in some area of scientific knowledge -- in this case, evolution -- they are to be filled by God. It is a "theory" that admits that evolution and natural selection explain such things as the development of drug resistance in bacteria and other such evolutionary changes within species but also says that every once in a while God steps into this world of constant and accumulating change and says, "I think I'll make me a lemur today."

But that definition of ID is simply incomplete. ID is at its essence an attempt to falsify the theory of evolution, not a replacement theory of the origins of life and speciation. You can accept ID as a challenge to evolution without adopting a supernatural replacement theory. However, the scientific establishment is not exactly eager to accept ID on these terms because it has no naturalistic alternatives to evolution. To understand ID you must first understand that the theory of evolution depends on two key elements: Genetic mutation and the process of natural selection. Genetic mutations are the randomizing change element. Take them away and you have an endlessly repeating, static gene pool. Natural selection pushes random change in the direction of recognizable speciation. It does this by encouraging the replication of genetic mutations that provide organisms an adaptive advantage. Remove natural selection and the gene pool would seem to churn randomly with no discernable, sustained direction. Intelligent design simply identifies specific biological structures (or processes) for which natural selection could not have provided directional impetus because a) those structures require multiple evolutionary steps to exist; and b) none of the intermediate steps in the evolution of the structure could have provided an adaptive advantage. A key concept in understanding ID is "irreducible complexity". Something is irreducibly complex if it requires all of its parts to be useful at all. The most common non-biological example is the traditional, spring-loaded mousetrap. If you remove any element of the trap it becomes a pile of useless parts, not a less-effective mousetrap and not particularly useful for other purposes. Common examples of irreducibly complex biological structures include the bacterial flagellum and blood coagulation. Remove natural selection as an influence in the formation of an irreducibly complex biological structure and you're arguing that several completely random genetic mutations either occurred at the same time in the same organism or persisted in the same part of the gene pool for millions of years until they were all present to provide an adaptive advantage. The probability of these sequences of mutations occurring completely randomly is effectively zero. The name "intelligent design" derives from the claim that it takes a designer to account for the formation of these irreducibly complex structures. But whether you take that step or not the challenge of irreducible complexity to evolution must be answered. Krauthammer says ID is not scientific because it's not disprovable, or falsifiable. Using his definition of ID that may be true. But in truth there are at least two ways to rebut ID:
  • Demonstrate that structures claimed to be irreducibly complex are not, that the individual component parts are adaptively useful in some way
  • Propose a plausible alternative mechanism for their formation that doesn't depend either on outside intervention or natural selection
It's late so I'll leave it at that for now. I'm sure the argument over ID won't end any time soon. :)

Sunday, November 20, 2005

Skins' Statistical Beating Continues

Washingon's ghastly fumble fortunes returned again today with an 0-for-4 recovery performance that was just bad enough to cost them yet another game. (If you have no idea what I'm talking about see this post.) Over the past two games against the Eagles and Buccaneers the Redskins actually covered a league average 50 percent of fumbles, but today's performance dropped the team to 9-of-39 on the season, just a bit above their nadir a few weeks ago. The Skins effectively lost control of their playoff destiny by losing to the Raiders. They're now the fourth ranked team contending for two NFC wildcard spots, one game behind 6-4 Atlanta and two games behind the 7-3 Giants and Bucs. Washington could perhaps afford one more loss and squeak into the playoffs, as long as it isn't during the three-game All-NFC East stretch to end the season. We'll see...
Current Washington Post Redskins stories: Korn: Norv Sticks Dan, Wilbon: Skins Suck, Porter is DC Local.

Saturday, November 19, 2005

Post Still Missing Skins Fumble Story

The Washington Post's Jason La Canfora manages to write a 900 word feature on the Redskins' turnover differential without mentioning their fumble recovery problems. See here, here, and here for details on Washington's poor fumble luck.

Confirmed Case of Mad Lib

Canadian Cynic didn't like my last post contrasting real political persecution in North Korea with the American left's paranoid fantasy version.

His response is--recounting an Iraqi crime during the 1991 invasion of Kuwait that proved untrue. Which is of course completely irrelevant and demonstrates my point that there's little chance of engaging in rational discourse with the angry left.

Cynic is skeptical because he feels the North Korean Christians who were killed demonstrated less refined spycraft than he would expect of the average Ludlum-tutored Westerner.

Perhaps the story is untrue. We may never know for certain. But even smart people make fatal mistakes. And everyone in the world is not as bright as Cynic.

If the report I described isn't well-documented enough for his tastes he might browse over to Amazon's Political Oppression & Imprisonment section where he'll find enough sobering material to occupy him for several years.

Sad as it is I'm not surprised by his response. Confronting the hard truth that you're the Veruca Salt of dissidentry can't be easy.

Open trackbacks from: Mudville Gazette.

Thursday, November 17, 2005

Popping Skulls

"Crushing dissent" is one of the nonsensical catch-phrases the angry left slings around rather freely these days when discussing the Bush administration. Attempting rational conversation with anyone who says such things is a rather futile pursuit and so I generally ignore them. Occasionally, though, it's good to highlight just how foolish and ignorant of the world such folks really are. The New York Sun runs a terribly sad story today which accomplishes precisely that. In 1996 North Korean soldiers building another empty highway for Kim Il Sung discovered a Bible and a list of 25 names while demolishing an empty house. Those on the list were identified as the members and leaders of a secret Christian church, which perhaps met in the house. Everyone on the list was immediately arrested at their workplaces and imprisoned. Several weeks later the parishoners were brought to the road construction site where their church had been discovered. An audience of spectators "arranged in neat rows" awaited them. Five of the Christians were identified as church leaders: one pastor, two assistant pastors, and two elders. These five were bound hand and foot and made to lie in front of an asphalt roller. As the 20 church members and other spectators watched, the five leaders were accused of being Christian spies, and ordered to deny their faith and pledge to serve only Kim Il Sung and Kim Jong Il. The five said nothing. "Some of the fellow parishioners assembled to watch the execution cried, screamed out, or fainted when the skulls made a popping sound as they were crushed beneath the steamroller."
Open trackback links from: The Political Teen, Stop the ACLU, Mudville Gazette, Basils Blog.

Rockefeller the Weasel

Senator Jay Rockefeller has been roundly ridiculed this week for his comments during an interview with Chris Wallace last Sunday.

Most of the mockery has focused on two answers, one where Rockefeller says he warned the governments of Saudi Arabia, Jordan, and Syria in early 2002 of President Bush's already-made decision to invade; and the other where he half-heartedly denies responsibility for his vote to authorize force against Iraq.

But there's another particularly weaselly aspect to his comments that I've not seen mentioned elsewhere. Rockefeller says he warned the Arab governments of Bush's intent to invade in January of 2002:

I took a trip by myself in January of 2002 to Saudi Arabia, Jordan and Syria, and I told each of the heads of state that it was my view that George Bush had already made up his mind to go to war against Iraq, that that was a predetermined set course which had taken shape shortly after 9/11.

A bit later in the interview he's challenged to take responsibility for the vote he cast in October of 2002 and says:
I'm responsible for my vote, but I'd appreciate it if you'd get serious about this subject, with all due respect. We authorized him to continue working with the United Nations, and then if that failed, authorized him to use force to enforce the sanctions.
His first answer fits nicely with the left's line that Bush decided to invade Iraq early on and then manufactured intelligience to support his aims. The second answer is the same excuse John Kerry used in the presidential debates last year: That he voted for the war resolution only to give Bush negotiating leverage with the UN and Iraq.

But Rockefeller says he knew nine months before his vote that Bush had made up his mind to invade! Like I said, weaselly.

Monday, November 14, 2005

Gruden Goes For Two: Foolish Folly or Genius Gamble?

Washington's painful one-point loss to Tampa Bay yesterday was about as gut-wrenching a seesaw battle as I've seen in a while, with huge momentum-changing plays every few minutes, four critical instant-replay reviews, and one unreviewable blown call.*

The most interesting sequence of the game came after the Bucs' final touchdown. With the score 35-34 in Washington's favor Tampa Bay opted to kick the extra point and tie the game. Washington's Shawn Springs and Walt Harris timed the snap and flew around the left and right ends to block the kick (Harris was credited). But wait: They both jumped offsides giving the the Bucs another chance from the 1.

Rather than kicking again Gruden deployed his heavy set and Mike Alstott plowed in for the score and the 36-35 lead. Or did he? Replays seemed to show Alstott's elbow down short of the goal line, and the replay official ducked under the hood to sort it out. Deciding there wasn't conclusive evidence to convict, er, overturn the line judge who called the score, he let the conversion stand. The Redskins received the kickoff, gained one first down to near midfield, and then ran four plays gaining zero yards, handing the ball and the game to the Bucs.

So was Gruden's call for two the best choice? The announcers certainly thought so, comparing his decision favorably with Dick Vermeil's touchdown call to win the game in last week's Kansas City vs Oakland matchup. I wasn't so sure.

In that instance the Chiefs elected to go for the win with five seconds on the clock, meaning Oakland had no chance to get the ball back and score. As Gregg Easterbrook pointed out last week in his NFL.com column, the Chiefs had probably an 80-90 percent chance or better of scoring from inside Oakland's 1, while kicking to send the game into overtime offered no better than a 50 percent shot at victory. In Sunday's game, however, the Redskins would get the ball back and only needed a field goal to recapture the lead. Gruden's call increased the Skins' chances of winning outright if Alstott failed, and didn't prevent the Skins from winning even if Alstott scored. On further review, it turns out that not only was Gruden's call probabilistically the right one, but he should have gone for two points and the win even before the Skins' offsides penalty moved the ball a yard closer. To calculate this I coarsely modeled the likeliest outcome paths with back-of-the-envelope estimates of the probabilities involved at each step.

One path, for example, was Bucs Go For 2/Bucs Score/Bucs Stop Wash. This is the path that actually occurred, and it happens be the single likeliest path out of the 16 paths I modeled.

Some other likely paths were Bucs Kick XP/Bucs Make Kick/Wash 50 Yd Dr/Wash 42 Yd FG and Bucs Kick XP/Bucs Make Kick/Bucs Stop Wash/Bucs OT Win or Wash OT Win. The least likely path was Bucs Kick XP/Bucs Miss/Bucs Onside Rec/Bucs 30 Yd Dr/Bucs Miss FG.

According to my model, the Bucs chances of winning were just 32 percent had they opted to kick the extra point and 55 percent by going for two (assuming an 80 percent chance of converting). If there had been no Skins penalty and you drop the Bucs chances of converting from the 2-yard line to 70 percent, that's still the better option with a 48 percent chance of ultimate victory!

The key is that by going for two the Bucs eliminate the outcomes where the Redskins have two opportunities to win. In other words, if the Bucs kick for the tie, Washington gets a chance to win in regulation (33 percent) and another chance of winning in overtime (50 percent). When the Bucs go for two they lose outright nearly 20 percent of the time but the other 80 percent of the time Washington just gets one 33 percent chance of victory. (Actually the Bucs lose outright 19 percent of the time--the other one percent they recover their onside kick, drive 30 yards with no timeouts, and kick a 42 yard field goal!)

UPDATE: I read today that NFL teams make just over 50 percent of two-point conversions from the 2-yard line. I'm not sure whether that number is accurate--it wasn't from an authoritative source. If it is, the sample size may be fairly small. I'd prefer to look at the percentage of all plays from the 2 which scored. But even assuming the worst-case 50 percent odds of success, going for two is still the slightly better option in this scenario, offering a 35 percent chance of winning vs a 32 percent chance of winning when teams kick the extra point!

One interesting side note on my probability estimates: I guestimated the average kicker's chance of hitting a 42 yard field goal at 65 percent. Ned Macey at FootballOutsiders happened to mention today that NFL kickers hit 62 percent of all 46-yard field goals. That was a pretty darn good guestimate! NFL.com stats so far this season show kickers hitting 69 percent of 40-49-yard field goals.

UPDATE II: More actual probabilities: On fourth-and-goal from the 1-yard line over the past four seasons teams score 67 percent of the time overall and 74 percent of the time when they run.

UPDATE III: I was asleep at the switch on this one (not paying much attention to the blog over Thanksgiving weekend) but thanks to Gregg Easterbrook for the link from his NFL.com column!


Current Washington Post Redskins stories: Skins Blame Selves, Wilbon, Wise, Kornheiser, Gruden: Easy Choice, Gibbs Gouges Refs, Skins Peek at Giants, Turner Still Stinks.
* Joey Galloway's out-of-bounds catch at Washingon's 3 in the first quarter was unreviewable because the official said Galloway was forced out of bounds by the defender.

Model Details

bucs-vs-skins-outcomes.gif (Click for larger Image.) Each step includes outcome descriptions, the probability of each outcome, and the number of outcome occurences. The orange rows indicate an ultimate Bucs win, the red indicate an ultimate Skins win. For example, here are the steps in the topmost path which starts when the Bucs Go For 2:

  1. Bucs score 80 percent of the time (16 of 20 outcomes on branch)

  2. Wash drives 50 yards 50 percent of the time (80 of 160 outcomes on branch)

  3. Wash kicks a 42 yard field goal 65 percent of the time (1040 of 1600 outcomes on branch)

  4. No more decisions on this path so Wash wins 100 percent of the time (20,800 of 20,800 outcomes on branch).