Thursday, August 5, 2010

On The Rate of Wrongful Conviction: Chapter 3.3

As I have mentioned eight times previously, I am preparing a monograph on the rate of wrongful conviction. Each chapter will deal with one estimate of that rate, beginning with zero and ending beyond 10%. I am posting the draft chapters here, as I write them.  I have so far posted the following: 

Chapter 0.027: The Scalia Number
Chapter 0.5: The Huff Number
Chapter 0.8: The Prosecutor Number
Chapter 1.0: The Rosenbaum Number
Chapter 1.3: The Police Number
Chapter 1.4: The Poveda Number
Chapter 1.9: The Judge Number
Chapter 2.3: The Gross Number

Now, for the last time, we come in beneath the 5% barrier. Hang in there while reading this longish post. There's good stuff at the bottom.

Chapter 3.3
The Risinger Number

In his paper “Innocents Convicted: An Empirically Justified Factual Wrongful Conviction Rate”, Michael Risinger claims to be the first person to determine a wrongful conviction rate by dividing a number of exonerations by an appropriate number of convictions.
To a great extent, those who believe that our criminal justice system rarely convicts the factually innocent and those who believe such miscarriages are rife have generally talked past each other for want of any empirically justified factual innocence wrongful conviction rate. This article remedies at least a part of this problem by establishing the first such empirically justified wrongful conviction rate ever for a significant universe of real world serious crimes: capital rape-murders in the 1980s.
Risinger then points out in a footnote that he wasn’t actually the first, having lagged Professor Ledewitz by 19 years.
Prof. Ledewitz recounted that he had examined the first hundred cases of execution under the restored death penalty and found four convicts (identified by name) whom he regarded as having “significant claims of innocence.” Again, the methodology is exceedingly soft, but the 4% resultant is not very different from that arrived at in this study.
Risinger also seems to be unaware or have lost track of Marty Rosenbaum’s 1991 paper “Inevitable Error: Wrongful New York State Homicide Convictions, 1965-1988”, the basis for Chapter 1.0 of this work.

Furthermore, Risinger seems to be unaware of or have lost track of Tony Poveda’s 2001 paper “Estimating Wrongful Convictions,” the basis for Chapter 1.4 of this work.

Finally, Risinger seems dismissive of the work of those who derive wrongful convictions rates from judge-jury agreement data, such as Professor Bruce Spencer of Northwestern University whose work will be discussed soon in Chapter 10.0.

So according to my counting, Risinger’s paper is no less than the fifth paper attempting to define an empirically-based wrongful conviction rate. Risinger’s paper is the first I have seen, however, to clearly describe his work as finding a proper numerator and denominator.
Using DNA exonerations for capital rape-murders from 1982 through 1989 as a numerator, and a 407-member sample of the 2235 capital sentences imposed during this period, this article shows that 21.45%, or around 479 of those, were cases of capital rape-murder. Data supplied by the Innocence Project of Cardozo Law School and newly developed for this article show that only two-thirds of those cases would be expected to yield usable DNA for analysis. Combining these figures and dividing the numerator by the resulting denominator, a minimum factually wrongful conviction rate for capital rape-murder in the 1980s emerges: 3.3%.
Unlike Samuel Gross, just discussed in Chapter 2.3, Risinger limits his analysis to rape-murder capital cases in which DNA turned out to be available post-conviction. Risinger also used a narrower range of years than did Gross, and together these differences presumably account for the difference in their estimates.

For his numerator, Risinger used the 11 DNA exonerations for that time frame reduced by one-half of an exoneration to account for the possibility that one in twenty of those people exonerated might be factually guilty.  For his denominator, Risinger used the 2235 capital sentences for that same time frame, multiplied by 21.45% to account for the fraction which were rape-murder convictions, multiplied by 2/3 to account for the fraction which had usable DNA for analysis. His math presumably looked something like:

(11 - 0.5) / (2235 x .2145 x .667) = 0.0328 = 3.3%

Risinger declared that number to be the lower limit for wrongful murder-rape capital cases, suspecting that DNA data had not been examined in all cases where it was available, but had been examined in at least 50% of those cases where it was available. He offered no data to back up those assumptions. He simply used them to argue the actual rate is closer to 5%:
It would be quite surprising, perhaps even shocking, if capital post-conviction counsel had failed to request DNA testing in anything close to half the 319 capital rape-murder cases in the reference set. Additionally, even in these cases, it is likely that the requests, if not universally made, would be skewed toward being made in the otherwise more factually questionable cases. So I believe we can conclude without much doubt that the ceiling is not double the floor (which would give a maximum ceiling figure for actual innocence of 6.4%), but is in fact substantially less. I believe it is fair to put a reasonable maximum under these circumstances at around 5%.
Risinger then discussed the possible extension of his number to crimes other than capital rape-murder cases.  For this he used words rather than a numerator and a denominator, and Ronald Allen called him on it in “Deadly Dilemmas.”
Michael Risinger's study examines error rates in capital rape-murder trials between 1982 and 1989. Risinger concludes that the error rate of false convictions ranges conservatively somewhere between 3.3% and 5%. The numerator comes from fairly convincing DNA exonerations in the set of capital rape-murder cases during the period, and the denominator is the set of capital rape-murder convictions during the same period. Risinger then considers extrapolating this error range to rapes, capital murders, and non-capital murders "manifesting particular callousness or brutality analogous to capitally sentenced murders," finding "no good reason for believing" that the rates would be much different.

But there are powerful reasons to think that the error rates might be different. Risinger's entire sample involved trials, yet most criminal charges are resolved by pleas. … Thus, an accurate error rate for capital rape-murder cases should reflect that data. Risinger makes the striking claim that the rates of error in plea-bargained cases could be as high as errors at trial. Such an unsubstantiated and highly improbable proposition stands in stark contrast to the commendably empirical cast to his article. Unless the frequency of false confessions is as high as the frequency of false convictions (and that appears implausible on its face), the error rate at trial cannot simply be hypothesized as the error rate of pleas.
“[Y]et most criminal charges are resolved by pleas.”  There’s the rub. Approximately 95% of all convictions result from plea bargains, not trials. That’s why so many prosecutors claim to have a 95% conviction rate. It has almost nothing to do with their performance before a jury.

So while the 95% plea bargain percentage works out fine for prosecutors, it is troublesome for those empiricists who estimate wrongful conviction rates from a small subset (DNA exonerations) of a small subset (capital rape-murder trials) of all convictions (most of which were plea bargains.) The plea bargain issue is of such magnitude that it places justice and injustice in the hands of prosecutors far more so than  in the hands of jurors. If there is a significant wrongful conviction rate in this country, it is due almost entirely to the failings of prosecutors.

Just as Michael Risinger dismisses the problem with a verbal slight of hand, so does Ronald Allen. Michael Risinger finds “no good reason for believing” the wrongful conviction rate of plea bargains would be much different than that of jury trials. Ronald Allen declares such a proposition “implausible on its face.”

Since Michael Risinger and Ronald Allen seem to be squabbling, and since the wrongful conviction rate for this country cannot be understood without understanding the wrongful conviction rate associated with plea bargains, I therefore hope, later in this work, to remedy “at least a part of this problem by establishing the first such empirically justified wrongful conviction rate ever for a significant universe of real world” plea bargains.

For now, I suggest the two combatants consider the instances of mass exonerations as evidence that innocent people are easily and frequently persuaded to confess and/or plead guilty to crimes they did not commit. Consider first the case of the Central Park 5.

The Central Park Jogger case involved the brutal assault, rape, and near murder of Trisha Meili. On April 19, 1989, the 28-year-old Meili was jogging through Central Park at night. She was attacked and left for dead, discovered 4 hours later suffering massive blood loss, internal bleeding, severe hypothermia, multiple lacerations, and a skull fracture so severe one eye was displaced from its socket. She was not expected to live, but she did. Today she still suffers from some loss of vision and balance, but has made a career as a motivational speaker.

After being arrested and interrogated, 5 black youths confessed to the crime, 4 of them confessed a second time while being recorded, each implicated the others, then all recanted. Though their confessions contradicted the forensics and the other confessions, and though DNA taken from the victim was from a single male other than any of those who had confessed, all 5 youths were convicted by juries, based primarily on their confessions.

In 2002, after the statute of limitations had expired, an inmate serving a life sentence for rape confessed to the crime, and he had the DNA to back it up. He was soon linked to 8 other rapes in a seven-month-period prior to the Central Park rape, including one in Central Park just two days earlier.

The 5 youths were all exonerated, though they had each already served their sentence. Despite Ronald Allen’s claim that it is "implausible on its face" that people would confess in substantial numbers to crimes they did not commit, all 5 confessed under conditions appellate courts found acceptable, conditions police defend as appropriate to this day.

If 5 false confessions in a single case are insufficiently plausible, consider the Tulia 42. Tulia is a small town in Texas not far from Amarillo. In 1999, authorities rounded up 46 citizens, 40 of whom happened to be black. Those 40 were represented approximately one-third of Tulia’s black male population. From Wikipedia:
All charges were based on the word of undercover officer Tom Coleman, a so called "gypsy cop" who made his living traveling through impoverished rural Texas offering to work undercover cheaply for short periods of time for under-funded police departments. Coleman claimed to have made over one hundred drug buys in the small town. He never recorded any of the sales, but claimed to have written painstaking notes on his leg under his shorts and upper arm under his shirt sleeve when nobody was looking.

During the roundup, no large sums of money, illegal drugs, drug paraphernalia, or illegal weapons were found. The accused drug dealers showed no signs of having any income associated with selling drugs. The drugs Coleman claimed to have bought from the accused did not have the fingerprints of the accused on them or their baggies. No independent witnesses could corroborate Coleman's claims. In his testimony, Coleman gave inaccurate descriptions of the "dealers" he had allegedly bought cocaine from. One suspect had his charges dropped when he was able to prove he had been at work during the times he had supposedly sold Coleman cocaine. Another produced bank and phone records indicating she was in Oklahoma City, Oklahoma at the time of her alleged crime. Many of the accused, however, seeing the long sentences dealt by all-White juries in earlier cases, pleaded guilty in return for lighter sentences, despite their proclaimed innocence. The remaining defendants were convicted solely on the basis of Coleman's testimony. John Cornyn, the state attorney general [and now a US Senator], awarded "Lawman of the Year" to Coleman.

Amarillo civil rights attorney Jeff Blackburn began investigating the Tulia defendants' cases along with civil rights organizations and a handful of attorneys from firms around the country. Eventually the case became a cause célèbre, and money was raised to legally challenge the cases. Many had already served several years in prison before this process gained momentum. By 2004, Blackburn and his team had freed most of the "Tulia 46" and a $6,000,000 collective settlement was reached to avoid further litigation in civil court.
More specifically, and more to the point being made herein, twelve of those arrested faced jury trials, were convicted and received sentences of 20 to 99 years. Thirty pleaded guilty out of fear of what jurors would do. As implausible as it may seem to Ronald Allen that anyone would plead guilty to a crime they didn’t commit, it seems that 30 out of 46 apparently did so in Tulia. That’s a 65% wrongful plea bargain conviction rate.
Ronald Allen might consider also the Dallas Sheetrock Scandal. At least 80 low-income, Mexican-American defendants were charged with possession of crack cocaine, based on the claims of a confidential informant paid $200,000 for his service to the community. Many of the defendants pleaded guilty before the cocaine was finally tested. It turns out that, by some strange coincidence, the cocaine in each case was not cocaine at all; it was gypsum powder, the primary constituent in sheetrock wallboard. Defense attorneys claimed prosecutors threatened to substantially increase the prison sentence being offered in return for a guilty plea if the defendant demanded the “cocaine” be tested.

In Hearne, Texas, the district attorney offered to drop charges pending against an informant if that informant would produce information leading to 20 drug arrests. The informant went above and beyond, providing information leading to the arrest of 28 people, all black, all for small-time drug sales. It all came unglued when one person refused to plead guilty and insisted on a trial. The district attorney dropped charges against that person and the other 16 who had not yet pleaded guilty. With respect to the 11 who had already taken a plea, the DA said “I don’t doubt one minute their guilt in dealing drugs.” The DA apparently finds it implausible that anyone would plead guilty to a crime they did not commit. Assuming the 11 who pled guilty were in fact innocent, as we have every reason to suspect, that’s 11 false pleas out of 11 convictions, a 100% false plea bargain conviction rate.

Out in Los Angeles, matters are no better. In September 1999, the Rampart Scandal exploded. From Wikipedia:
More than 70 police officers in [an anti-gang unit] were implicated in misconduct, making it one of the most widespread cases of documented police misconduct in United States history. The convicted offenses include unprovoked shootings, unprovoked beatings, planting of evidence, framing of suspects, stealing and dealing narcotics, bank robbery, perjury, and covering up evidence of these activities. … As a result of the probe into falsified evidence and police perjury, 106 prior criminal convictions were overturned.
Other estimates place the number of trumped up cases as high as 150. The great majority of those framed were young Hispanic men who ended up pleading guilty to false felony gun or drug charges.

I am not arguing that these cases of mass exoneration are representative of the criminal justice system. I hope they are rare exceptions. I do argue, however, they provide evidence that out of any select group of people, many can be compelled to confess or plead guilty to crimes they did not commit, as implausible as that may seem.

In summary, Michael Risinger calculated a wrongful conviction rate of 3.3% for capital rape-murder convictions. He then suggested that this rate might be applicable to other types of crimes, and perhaps even to plea bargains. Ronald Allen took scholastic umbrage and declared Risinger’s extension “implausible on its face.” I took blogger umbrage at Allen’s response. I offered Allen, my lastnamesake, examples of wrongful plea bargain convictions in an effort to convince him they are plausible. I made the bold claim that I will, in this work, become the first person to provide an empirically-based estimate of the rate of wrongful plea bargain convictions.

1. “Innocents Convicted: An Empirically Justified Factual Wrongful Conviction Rate,” by D. Michael Risinger was published in The Journal of Criminal Law & Criminology, Volume 97, Number 3, in 2007; Copyright © 2007 by Northwestern University School of Law

2. At the time “Innocents Convicted” was published, D. Michael Risinger was a Professor of Law at Seton Hall University School of Law. Since 2008, he has been the John J. Gibbons Professor of Law at Seton Hall University School of Law.

3. “An Essay Concerning Judicial Resignation and Non-Cooperation in the Presence of Evil,” by Bruce Ledewitz, 27 Duquesne Law Review 1(1988).

4. When I write of Ronald Allen, I refer to both Ronald J. Allen and Larry Laudan.

5. “Deadly Dilemmas,” by Ronald J. Allen and Larry Laudan, 41 Texas Tech Law Review 65 (2008).

6. At the time “Deadly Dilemmas” was published, Ronald J. Allen was the John Henry Wigmore Professor at Northwestern University School of Law, a position he still holds today.

7. At the time “Deadly Dilemmas” was published, Larry Laudan was Senior Investigator and the Instituto de Investigaciones Filosóficas, Universidad Nacional Autónoma de México.

8. Just as I thought Samuel Gross’s 2.3% number is far too low, I believe Michael Risinger’s number is far too low. I will make my case in Chapter 12.6.

No comments:

Post a Comment