EvidenceProf Blog

Editor: Colin Miller
Univ. of South Carolina School of Law

A Member of the Law Professor Blogs Network

Monday, April 9, 2012

Julia Rickert's Denying Defendants the Benefit of the Doubt & an Argument for Empiricism in Rule 609 Impeachment

Late one cold night in Chicago, a homeless man came upon an unlocked car parked on the street. He decided he would sleep in it. Early the following morning, he awoke just as a police cruiser pulled parallel to the car. He was arrested and later charged with burglary. Because of his criminal history, the man faced six to thirty years if convicted. The prosecutor offered him eight years in exchange for a guilty plea.

The crime of burglary, a Class 2 felony in Illinois, is committed when one "knowingly enters or without authority remains within a building, housetrailer, watercraft, aircraft, motor vehicle as defined in The Illinois Vehicle Code, railroad car, or any part thereof, with intent to commit therein a felony or theft." The requisite intent to commit a felony or theft can be inferred from the bare circumstance of having entered without authority. This inference can of course be rebutted.

In this case, the defendant had not disturbed any of the valuables in the car. The police had been called by the car’s owner, who reported the presence of someone in his vehicle, but did not specify whether the person was awake or asleep. The police report did not comment on whether the homeless man appeared to have just awoken, but it did indicate that sunglasses were found in the homeless man’s pocket. The vehicle owner told the police that the sunglasses looked familiar and may have been left in his car by a friend. The trespasser, however, claimed the sunglasses were his own. What no one disputed was that the car contained items of value that had not been disturbed, such as a cellular phone, a stereo, and compact discs.

If this defendant is charged with burglary and wants to argue that he is only guilty of trespass, how is he going to rebut the inference of intent? The obvious answer is "by testifying." But what if the defendant has two prior convictions for criminal sexual assault. Would evidence of these convictions be admissible in the event that the defendant testifies, in effect putting him between a rock (not testifying) and a hard place (testifying and having the jury hear about his prior convictions)? This is the question addressed by Julia Rickert, currently a clerk for the United States Court of Appeals for the Seventh Circuit, in her article, Denying Defendants the Benefit of the Doubt: Federal Rule of Evidence 609 and Past Sex Crime Convictions, 100 J. Crim. L. & Criminology 213 (2010).

The hypothetical that led this post is the hypothetical that led Rickert's article, and it is taken from a real case that she observed while a summer clerk for the Cook County Public Defender. The difficulty for a court in deciding whether to admit the defendant's prior convictions in the event that he testifies is that it must balance probative vale against prejudicial effect under Federal Rule of Evidence 609 (or, in the hypothetical, Illinois Rule of Evidence 609).

The probative value of the convictions consists of how much those convictions tell us about the likelihood that the defendant will perjure himself on the witness stand. If they tell us a good deal (as with most property crimes), they are highly probative. If they don't tell us much (as with most crimes of violence), they are minimally probative. Meanwhile, the potential prejudicial effect twofold: (1) We don't want the jury to use the prior convictions to conclude, "Once a sexual predator, always, a sexual predator" or, more generally, "Once a criminal, always a criminal; and (2) We don't want the jury to think that even if there is insufficient evidence to prove burglary, it wouldn't be a big deal to convict the defendant of the crime because it is not the worst thing in the world to take a two time sex offender off the streets for an extended period of time. By looking at the hypothetical above, you get a good sense of the general arguments that Rickert makes on both sides of the issue, and I would certainly recommend that you read her terrific piece for a full explication of these issues. 

In this post, though, I want to focus on one statistic I that Rickert cites that I had seen before but which I had never realized has relevance to Rule 609. According to Rickert,

studies show that those who are convicted of sex offenses are significantly less likely to commit non-sex crimes than are other types of criminals.[FN61] Does this mean that sex offenders are less likely to commit the crime of perjury than are defendants previously convicted of non-sex crimes? Perhaps not, but in any event, I have encountered no argument that past sex crimes are especially probative of a witness’s propensity to lie on the stand.

[FN61] "Sex offenders were less likely than non-sex offenders to be rearrested for any offense—43 percent of sex offenders versus 68 percent of non-sex offenders." BUREAU OF JUSTICE STATISTICS, U.S. DEP’T OF JUSTICE, available at http://www.ojp.usdoj.gov/bjs/crimoff.htm#recidivism (last visited Oct. 26, 2009). This statistic should not be read as suggesting that sex offenders don’t have a tendency to commit future sex offenses: "Sex offenders were about four times more likely than non-sex offenders to be arrested for another sex crime after their discharge from prison—5.3 percent of sex offenders versus 1.3 percent of non-sex offenders." Id. Furthermore, "Within 3 years of release, 2.5% of released rapists were rearrested for another rape," while only "1.2% of those who had served time for homicide were arrested for a new homicide." Id.

In other words, while statistics can only tell us so much, the above statistics strongly imply that convicted sex offenders are much less likely than most other witnesses with past convictions to perjure themselves on the witness stand. And I think this makes intuitive sense. A person who commits a sexual offense is a sexual deviant, and there is little reason to believe that this sexual deviancy would lead the person to commit perjury. It is this deviancy and not (necessarily) a disregard for society's laws that leads to the crime. Conversely, someone who commits larceny, drug offenses, or even a violent crime such as battery shows a certain disregard for obeying society's laws and thus has a higher likelihood of committing perjury than the sex offender. Therefore, there is a decent argument that convictions for sexual crimes should be the least probative convictions under Rule 609, rendering them rarely admissible.

That's the smaller point, and the more general point is that perhaps courts should start engaging in a more empirical analysis of what types of prior convictions predict perjury. My sense from reading gobs of Rule 609 cases is that judges shoot from the hip in deciding, say, that a larceny conviction has decent bearing on witness honesty and that an assault conviction does not. But what do the statistics actually show? If we looked at all defendants convicted of perjury in the past decade, what percentage of these defendants would have larceny convictions? What percentage would have assault convictions? And what percentage would have convictions for sexual crimes? I'm not sure that a court, litigant, or scholar has ever done such an analysis, but it is sure something that could increase the efficacy of Rule 609.



| Permalink

TrackBack URL for this entry:


Listed below are links to weblogs that reference Julia Rickert's Denying Defendants the Benefit of the Doubt & an Argument for Empiricism in Rule 609 Impeachment:


In addition to least probative, prior sex offenses may be among the most highly predjudical and most regularly permitted on cross, even under Rule 609 and it's reverse balancing test for testifying defendants.

Posted by: Wes Porter | Apr 10, 2012 6:36:26 AM

As a (mostly criminal law) practitioner in Alaska, I am thankful that our state Evidence Rule 609 is limited in the following way: "[E]vidence that the witness has been convicted of a crime is only admissible if the crime involved dishonesty or false statement."

Even then, the effect of impeachment can be minimized, as a colleague of mine has noted. In a domestic violence trial when his client was testifying: The prosecutor asked "You have been convicted of a crime of dishonesty, correct?" "No," the defendant answered.
P: In 2002, you were convicted of larceny, correct?
D: Yeah
P: So you agree that you've been convicted of a crime of dishonesty then.
D: No. That's not lying, that's stealing.

Posted by: Kevin | Apr 10, 2012 9:46:42 AM

I'm not sure that empiricism gets us very far, though. The issue isn't whether or not someone like the defendant is going to lie but whether that specific defendant is going to lie. To my mind, I think the proper rule is the one that is present in Alaska. The idea that someone who steals is more likely to lie makes no logical sense as a matter of categorization. Trying to correlate sex offenses and lying is comparing apples and cars.

Posted by: Daniel | Apr 10, 2012 10:25:36 PM

In Washington State, the rules of evidence convictions over 10 years old cannot be used as 'prior bad acts' UNLESS they were crimes of 'moral turpitude'. So conviction of, say, welfare fraud 22 years ago would still debilitate a defendant from being able to testify in their own behalf? If so, my remedy would be to argue the prosecution has to reach back 22 years to attack the defendant's character/testimony because they have insufficient facts/evidence to convict.

Posted by: pinbalwyz | Apr 13, 2012 10:02:21 AM

Post a comment