a public defender

If you're new here, you may want to subscribe to my RSS feed. Thanks for visiting!

Grits for Breakfast points us to this important and disturbing story published in the L.A. Times a few days ago. It is a report on the murder of 16-year old Martha Puebla, whose name the police used while fabricating an identification.

They were trying to get her boyfriend for an unrelated murder and during their interrogation of him, they showed him a photo array (already a source of many problems) where they forged a circle around his picture with Martha’s initials and an “identification” beneath it.

To drive home his point, [police officer] Pinner laid down a “six-pack” — an array of mug shots that detectives often show to witnesses or victims of crimes. On it, [suspect] Ledesma’s photo was circled, and the initials “M.P.” were written below it. “Those is the guy that shot my friends boyfriend” was scrawled along the margin, followed by Puebla’s signature.

“I don’t even know a Martha,” Ledesma lied.

Police deceit during investigations and interrogations has long been tolerated (see, e.g. Illinios v. Perkins), but this may be one of those instances of the disconnect between theory and reality. In this case, it lead to Puebla’s death:

The next night, Ledesma reached for a pay phone outside his cell. “Cokester,” he said into the receiver, calling his friend Javier Covarrubias by one of his gang monikers, “do you know the slut that lives there by . . . my house? Her name starts with an M . . . I need her to disappear. She is dropping dimes.”

To the gang, Puebla was a snitch and needed to be dealt with.

“Uh huh, like that,” Ledesma told Covarrubias, using a mix of Spanish and English. “But [keep a] low-pro[file]. ..Stay on your toes, homie. And don’t get caught.”

Of course, this is also a product of the anti-snitching culture that has garnered much attention. But the fact remains that as a result of the deceit on the part of these law enforcement officers, a young girl is dead.

Cases like Illinois above have given cops free reign of the interrogation room - everyone’s heard of good cop, bad cop - and license to lie about almost anything, resulting in false confessions.

While I recognize the arguments supporting lying to suspects during interrogations, there has to be some sort of oversight and limits placed on the extent of permissible deceit. Certainly, endangering the life of someone who is innocent and whom the cops should know would be in danger of losing his/her life as a result of their lie should be outside those limits.

The responsibility of law enforcement is to protect citizens and ensure their safety, not to endanger them while engaged in a tunnel-visioned pursuit of “catching a criminal”.

So what is the line? When is deceit okay, if at all? Should it hinge on waiver of Miranda rights? I’ll expound on those thoughts in a later post.

Mike at C&F reviews a book that he recently read (and recommends), entitled “Don’t Believe Everything You Think: The 6 Basic Mistakes We Make in Thinking.” It is primarily a book about the human mind and the tricks it plays on us. Prof. Greenfield astutely reminds us all not to fall into these traps, if we are to represent our clients adequately.

I looked at it from another perspective. I think, generally speaking, as defense attorneys we are aware of these “tricks”. Some of them read just like the problems with eyewitness identification that are the subject of many papers and lots of research.

But this “list” organizes them succinctly (which, incidentally, is one of the “tricks”). If this is truly how people think, then we have just gained an insight into the minds of our jurors. We need to be aware of these “tricks” and be ready to either use them or to expose them.

Consider the following, with Mike’s summary:

Confirmation bias. People hate being told they are wrong. People love being told they are right. Consequently, people only look for data proving themselves right. Yet, in so doing, who knows what evidence we are missing proving ourselves wrong?

Anyone who is up on eyewitness id research knows about this. Witnesses reinforce their own memory and come to believe very strongly that what they saw is what they recollect. (As a side note, this is why it is imperative in eye id cases to attempt introducing the testimony of an expert.)

We are super simple Simons. Why do stock markets rise and fall? Why did your wife leave you? We have one or two sentence explanations of very complex events. In a sense, our need to oversimplify stems from the narrative fallacy and our inability to appreciate chance. Give us a show story that makes the world seem logical, and we’re sold.

Because of that, we prudently invest in mutual funds and worry about the child molester next door rather than the one in our own homes.

This, I think, can work in two ways. Find that two sentence theory of your case and you might be able to convince the jury of whatever your position is. Again, this is not new, but just further reinforcement that you want to keep it as simple as possible. One of the first things I learned about trying a case to a jury was to keep is as simple as possible - and simple to us lawyers is not the same as simple to lay jurors. Break the case into small, bite-sized pieces and feed as few of them as possible to jurors.

The other way it can work is to understand that the State’s case is usually made up of such a simplicity. A man saw the defendant hit someone. Hence, the defendant is guilty. Knowing that people lean toward such simple solutions teaches us that we need to unravel them carefully, and simply. Take each strand, one at a time and deconstruct it.

The simpler the better. See Occam’s Razor (or Bennett’s Chainsaw). I’ll have to read the book, but in the meantime feel free to jump in with your thoughts on this.

Just to show you how skewed the priorities of legislators are, let us compare two bills side by side.

On the left, we have the eyewitness id reform bill and on the right, the “collect DNA from the innocent” bill.

One is clearly needed, the other could be a significant violation of due process. (Yes, I do know that some states have held such DNA collection to not be a violation of the 4th. I disagree with them.)

As EyeID points out, the bill, while missing some legs, would have been a significant step in the right process. Alas, it was not to be. However, the esteemed legislators have deemed it worthy to collect DNA samples from people arrested of crimes. The argument behind this piece of legislation is that collection of DNA at the time of arrest would permit law enforcement to solve more crimes…because, you know, if you’re arrested, then you’ve probably committed a crime in the past.

Another rationale put forth by the State [pdf]- and I do love this - is that collection of DNA at the time of arrest would not only serve to solve unsolved crimes, but also prevent wrongful convictions.

The irony here, lest it be lost on you, is that the very same State opposed [pdf] the eyewitness ID reform bill, when DNA exonerations have shown that in 75% of wrongful conviction cases faulty IDs have been the culprit. Apparently, in eyewitness ID reform, there is a “pilot program” and the “jury is still out” on whether sequential or simultaneous lineups are better. Yes, you read that right. Apparently, no one knows how to use Google. There is some heavy citing of the Mecklenburg Report, which permits law enforcement to continually bury its head in the sand. The State’s testimony then quotes Gary Wells out of context (I’ve read about this being done in other states too) and ignores his response to this misquoting (I’m going on memory here - I’m sure the guys at EyeID know what I’m talking about - or if I’m imagining this whole thing, I’ll take it down).

The CT Public Defender’s response to the EyeID bill is here and to the DNA bill here. All the testimony submitted on these bills is available here.

On a positive note, the committee did pass the probation reform bill, which I discussed previously.

All the bills reported out of committee by last night’s deadline are here. For example, here’s a bill “encouraging” bar owners to install breath alcohol testing devices. Here’s a bill making it illegal to hang on a noose on public property, or private property without the consent of the owner.

Here’s a legal fiction that we live with: Defendant is accused of murder. Defendant is on probation while he is alleged to have committed this crime. The state presents an eyewitness to the crime. The jury disbelieves the eyewitness and returns a verdict of not guilty. Later, relying on that very same evidence, a judge finds the defendant guilty by a preponderance of the evidence of violating his probation and sentences him to 8 years imprisonment.

What is wrong with this picture? Legally, there is nothing wrong. There are different standards of proof. A jury must find a defendant guilty beyond a reasonable doubt and a just must only find that the defendant violated his probation by a preponderance of the evidence.

In a case like this, however, it is interesting to analyze this further. All that can be gleaned from the news story about the state’s case is that it turned on the testimony of the eyewitness. The jury disbelieved the eyewitness, finding her account incredible. For the jury, there were only two options: either they believed her or they didn’t. They didn’t go back to their deliberating room and decide that they “didn’t believe her beyond a reasonable doubt”. Normal people don’t think like that; lawyers do. The jury certainly wasn’t sitting in that room saying: “well, it’s more likely than not that she was telling the truth, but it isn’t likely beyond a reasonable doubt”.

So they disbelieved the eyewitness. Zero credibility. They acquitted.

Yet, the judge, as was his right, believed the witness. Again, I don’t think a judge is sitting there thinking “It is more likely than not that she is telling the truth”. Either you believe the witness or you don’t. (I understand that you may believe parts of the witnesses testimony, but the jury clearly didn’t believe the part that had the defendant committing the crime and yet the judge did.)

So the judge is essentially telling the jury: “You got it wrong. I, one person, am right and you, six people, are wrong”. He’s allowed to do that.

Essentially, the man was acquitted and convicted based on the exact same evidence. Regardless of standards of proof, that should not be allowed to stand.

Via CDW comes this fantastic new paper by Jules Epstein, which examines the problems with eyewitness identification testimony and the short-comings of using cross-examination to challenge it. It is a must read for the practitioner. The first 40 pages or so trace the history of eyewitness identification and of cross examinations and their place in our adversarial system. Then it underscores the need for expert testimony in eyewitness ID cases by shattering the myth propagated by judges and appellate courts that cross-examination will bring out any untruths.

That is because, often, eyewitness testimony does not contain untruths. The problems associated with eyewitness testimony are such that it is nearly impossible to expose them on cross.

1. Cross racial IDs: How does one go about questioning a witness regarding this sensitive issue, which has been demonstrated to be a serious problem in identifications? One cannot simply ask a witness if he/she is better at identifying people of their own race or if they are aware that studies show that such a bias exists.

2. Weapons focus:

The entire premise of weapons focus is that it is often a subconscious phenomenon—without realization of the occurrence, the witnesses’ eyes are drawn toward the weapon. It is precisely the extent to which the witness is unaware of the diverted attention that cross-examination proves ineffective.

This might be the only area where it is possible to do something on cross. As the example in the paper illustrates, the cross can elicit significant details about the weapon, thereby proving (or sowing seeds of doubt) that the witness was not focused on the face, but rather on the weapon.

However, the problem still remains that many jurors believe that a weapon increases attention overall and makes the eyewitness more reliable.

3. Stress: One can easily prove the fact of stress, but it is almost impossible to prove the impact or consequence of stress via cross. These are scientific results and ideas that cannot be elicited through the lay witness on the stand and often-times, the witness will use stress to affirm their recollection.

4. Memory Retention and the Confidence-Accuracy disconnect: This is another one that’s impossible to establish on cross. Asking a witness whether their memory has gotten worse over time and that just because they think they’re right doesn’t mean they’re right will result in them simply re-affirming their identification.

5. System variables (sequential lineup, double-blind, etc.): What can be established via cross is the occurrence of imperfect ID procedures, but not the significance, as with stress above. So the witness was not told that the suspect may or may not be in the lineup. What does it mean to the reliability of the ID? How are you going to get that out on the cross of anyone, including the cop?

So what is to be done? One method - the example used in the paper - is the one I prefer. To go over the events in a chronological order, breaking it down into tiny, tiny bits. One piece of information at a time.

Of particular importance is the technique of “time-framing”—the art of breaking the event or crime into a series of discrete acts, each in isolation.

I’d like to hear from you, my practitioner reader. What have you found useful? Has anything worked at all? I seriously doubt we’ll ever get the “aha!” moment during the cross of an eyewitness.

What I think this paper does is gives us a roadmap to arguing the admissibility of expert testimony. The offer of proof is one thing, but setting up why it is necessary goes a long way to informing the judge that he/she should admit the testimony. This paper lays out all the reasons why it is necessary to inform the jury of the pitfalls of eyewitness testimony. Use it. Even if you don’t get the expert testimony in, it gives you leverage to argue to the judge that you need to either ask jurors about it or be able to argue some of it in closing and have the judge give a detailed instruction on the fallibility of eyewitness testimony.

I’ve uploaded it here [pdf]. It’s available for free from SSRN, so I figured I could make it available here too. If that’s a problem, someone let me know and I’ll take it down.

(Courtroom sketch: Wired News/Norman Quebedeau)

Fantastic news today. A new bill has been introduced to reform ID procedures in CT.

The bill provides for most of the recommended changes to ID procedures during lineups. I have reproduced the bill in its entirety below. Spread the word:

(b) Not later than January 1, 2009, each municipal police department and the Department of Public Safety shall adopt procedures for the conducting of photo lineups and live lineups that comply with the following requirements:
(1) When practicable, the person conducting the identification procedure shall be a person who is not aware of which person in the photo lineup or live lineup is suspected as the perpetrator of the offense;

(2) The photo lineup and live lineup identification procedures shall be conducted in sequence so that the eyewitness is shown each photograph or each person one at a time rather than viewing the photographs or the persons simultaneously;

(3) The eyewitness shall be instructed prior to the identification procedure:
(A) That the perpetrator may not be among the persons in the photo lineup or the live lineup;
(B) That the eyewitness should not feel compelled to make an identification;
(C) That each photograph or person will be viewed one at a time;
(D) That the photographs or persons will be displayed in random order;
(E) That the eyewitness should take as much time as needed in making a decision about each photograph or person before moving to the next one; and
(F) That all photographs or persons will be shown to the eyewitness, even if an identification is made before all have been viewed;

(4) The photo lineup or live lineup shall be composed so that the fillers generally fit the description of the person suspected as the perpetrator and, in the case of a photo lineup, so that the photograph of the person suspected as the perpetrator resembles his or her appearance at the time of the offense and does not unduly stand out;

(5) If the eyewitness has previously viewed a photo lineup or live lineup in connection with the identification of another person suspected of involvement in the offense, the fillers in the lineup in which the person suspected as the perpetrator participates shall be different from the fillers used in any prior lineups;

(6) At least five fillers shall be included in the photo lineup and at least four fillers shall be included in the live lineup, in addition to the person suspected as the perpetrator;

(7) In a photo lineup, no writings or information concerning any previous arrest of the person suspected as the perpetrator shall be visible to the eyewitness;

(8) In a live lineup, any identification actions, such as speaking or making gestures or other movements, shall be performed by all lineup participants;

(9) In a live lineup, all lineup participants shall be out of the view of the eyewitness at the beginning of the identification procedure;

(10) The person suspected as the perpetrator shall be the only suspected perpetrator included in the identification procedure;

(11) Nothing shall be said to the eyewitness regarding the position in the photo lineup or the live lineup of the person suspected as the perpetrator, except as otherwise provided in subparagraph (D) of subdivision (3) of this subsection;

(12) Nothing shall be said to the eyewitness that might influence the eyewitness’s selection of the person suspected as the perpetrator;

(13) If the eyewitness identifies a person as the perpetrator, the eyewitness shall not be provided any information concerning such person prior to obtaining the eyewitness’s statement that he or she is certain of the selection; and

(14) A written record of the identification procedure shall be made that includes the following information:
(A) All identification and nonidentification results obtained during the identification procedure, signed by the eyewitness, including the eyewitness’s own words regarding how certain he or she is of the selection;
(B) The names of all persons present at the identification procedure;
(C) The date and time of the identification procedure;
(D) The order in which the photographs or persons were displayed to the eyewitness;
(E) In a photo lineup, the photographs themselves;
(F) In a photo lineup, identification information and the sources of all photographs used; and
(G) In a live lineup, identification information on all persons who participated in the lineup.

One can only hope that this passes.

HT: CTLP

Apparently, scientists have developed a new tool to “freeze” crime scene memories.

The tool - a self-administered interview applied by witnesses at crime scenes - combats natural memory decay by using the latest research in cognitive psychology techniques. It ‘freezes’ images and details of crime scenes and perpetrators in the minds of witnesses, particularly small and seemingly insignificant details that provide major leads for detectives that turn out to be crucial in solving cases.

While this might have utility in memorizing details from the scene itself, I have to question its usefulness in remembering descriptions of perpetrators. Part of the problem with eyewitness identification testimony (and partly why experts are starting to be used) is eyewitness confidence. Studies have shown that there is very little correlation between eyewitness confidence and accuracy. I fear that this tool might serve to cement incorrect recollections of the perpetrator.

It is a tool that seems to work, though:

Tests at simulated crimes scenes were remarkable with witnesses using the tool recalling forensically relevant information 42 percent more accurate than other witnesses who were simply asked to ‘report as much as you can remember’. The tests also revealed the witnesses using the self-administered interview (SAI) were 44 percent more correct with details about people - therefore, possible suspects - who had been involved in the event.

In another test there was a delay of seven days between witnessing the event and providing a full account. Half the participants completed self-administered interviews after witnessing the event while the other participants simply gave their name and contact details - as normally happens to a large number of witnesses at crime incidents. Scientists tested the group after seven days and found participants who had completed the SAI were still reporting almost 30 per cent more correct details than other witnesses.

That’s just staggering. This goes to show that in every case we have involving eyewitness testimony, we must explore any and all challenges to its reliability and perhaps retain an expert. This cannot be ignored any more in practice. For the CT practitioner, Lisa Steele’s Law Review article at 25 Quinnipiac L. Rev. 799 is very helpful (thanks to JC).

As the dust on the 208th DNA exoneree had barely begun to settle, Broward County public defender Howard Finkelstein sent a letter to law enforcement officials suggesting a change in identification procedures.

Bostic’s [the exoneree] accuser recently told an investigator she never saw her rapist. She picked Bostic out of a photo lineup, she said, because she had seen him in the neighborhood in the days before the attack.

Simple extra precautions could keep this from happening again, Finkelstein said.

“These procedures will impact the human cost of misidentification,” he said. “This isn’t about pointing the finger at law enforcement. This is about making sure the methodology and the systems we employ are designed so innocent people don’t get ensnared in our system.”

Currently, Florida uses the non-blind, non-sequential method of identification. Finkelstein called for them to use the double-blind, sequential method. Law enforcement’s response was curious, if not typical:

“If we had concerns about the procedure, we would have changed the procedure,” said Elliot Cohen, spokesman for the Broward Sheriff’s Office. “But new ideas and new proposals are always worth looking at, and we’ll take it in that spirit.”

At least eyeid reform seems to be gaining some momentum. 16 states have considered some legislation in this regard during the past year. Connecticut, although one of those sixteen, couldn’t get past simply funding a pilot program. I’m not even sure that the pilot program has gone into effect.

EyeID does a terrific job (as always) with this post about the Telfaire instruction and the problems associated with it. The post then goes on to suggest what an appropriate instruction should contain:

These tailored instructions should include, where applicable, instructions on the cross-race effect, the detrimental effects of stress on eyewitness memory, the weapon-focus effect, the absence of a reliable correlation between confidence and accuracy, and lineup procedures that have been shown to make an identification more or less reliable.

But scientifically accurate jury instructions are not enough. To properly sensitize jurors to the problems with eyewitness testimony and to begin to curb the nationwide wrongful conviction problem due to over-reliance on bad eyewitness evidence, expert testimony is critical. In fact, according to the recent Copeland ruling by the Supreme Court of Tennessee, expert testimony is the only reliable method to ensure that jurors have the necessary tools to assess eyewitness evidence accurately. State v. Copeland, 2007 WL 1498396, *11 (”Research over the past 30 years has shown that expert testimony on memory and eyewitness identification is the only legal safeguard that is effective in sensitizing jurors to eyewitness errors.”).

Make sure you read the entire post.

I’ve been remiss in not mentioning EyeID’s terrific coverage of the debunking of a highly-publicized study out of Illinois that claimed to have raised questions about some of the procedural reforms being adopted around the country to improve the handling of eyewitness evidence.

The study, taking the form of a pilot project spearheaded by Chicago police across three counties, purported to reveal that current identification procedures protected against mistaken identifications better than reforms (PDF) proposed by respected social scientists, based on extensive research on eyewitness memory. In other words, the Chicago police were happy to report that, notwithstanding the 19 wrongful convictions recorded in Illinois that resulted from faulty eyewitness evidence, everything was just fine and no pesky (scientific) reforms were needed.

Two immediate red flags: The report was not subjected to peer review and it was authored by a lawyer for the Chicago Police Department. EyeID notes sadly that the study’s findings “were trumpeted on the front page of the New York Times and have since served as serious impediments to reform in Legislatures around the country.”

Then some scientists decided to take a look. They released their own analysis [pdf] of the CPD report and found

“the design [of the Illinois study] guaranteed that most outcomes would be difficult or impossible to interpret,” and the study’s fundamental flaw has “devastating consequences” with respect to its scientific merits.

Then the Chicago Tribune runs this curious story, suggesting that the best lineup format has yet to be identified. This is utter nonsense. As EyeID rightly notes,

The first problem is that the “disagreement” is one of conflicting agendas, not unresolved questions of science. Scientists, with their well-known bias in favor of facts, are in general agreement that sequential lineup procedures are less likely to put innocent people in prison. Every peer reviewed study on the topic bears out this finding. A meta-analysis (PDF) was conducted by leading psychologists to extrapolate the comparative accuracy rates of the two types of procedures, and the clear finding was that sequential lineups are far less likely to result in an innocent person being identified. When “moderator variables” are considered, the two methods are also largely equivalent in their likelihood of bringing about a correct selection of the actual perpetrator, when he is present in the lineup. The general consensus among scientists is that “sequential lineups are superior.”

To the extent that there is disagreement on this point, it is not a disagreement among scientists or any sort of conflict in empirical findings. Rather, it is an objection by prosecutors at the audacity of scientists who dare to suggest that science should inform criminal justice policy. After all, the Chicago Police Department has been doing lineups for decades. Who are these scientists to tell them their procedures are flawed? It’s like a bunch of M.D.s telling cigarette manufacturers that smoking causes cancer.

While the Illinois study may have stalled ID reform in some states, or made others question the validity of their pilot programs in the short term, it has also resulted in an explosion of interest in the area of ID reform. This can only be a good thing. As more and more people become aware of these proposed reforms and there are more studies confirming their effectiveness, states will have to take notice and ID reforms will gain more acceptance. As the study I reported about earlier noted, “the leading cause of the wrongful convictions was erroneous identification by eyewitnesses, which occurred 79 percent of the time.” It should be in everybody’s interests to significantly reduce that number.

See, I told you there was so much I had missed in just one day.

EyeID points to this article about two upcoming studies on wrongful convictions:

The first, “Judging Innocence”, is soon-to-be-published in the Columbia Law Review, conducted by University of Virginia Professor Brandon Garrett. Professor Garrett’s study systematically examined all of the DNA exonerations and concluded that “the leading cause of the wrongful convictions was erroneous identification by eyewitnesses, which occurred 79 percent of the time. In a quarter of the cases, such testimony was the only direct evidence against the defendant.” Other leading causes of wrongful convictions were faulty forensic evidence, unreliable snitch testimony and false confessions.

You must read the study [pdf] to get the full sense of the various problems found. Here’s a sample:

There were false confessions in 16 percent of the cases, with two-thirds of those involving defendants who were juveniles, mentally retarded or both.

Here’s the zinger:

Garrett’s study strongly suggests, then, that there are thousands of people serving long sentences for crimes they did not commit but who have no hope that DNA can clear them.

The second is by Professors Samuel Gross at Michigan Law School and Barbara O’Brien of Michigan State. It, too, reached similar conclusions. Specifically:

“The main thing we can safely conclude from exonerations is that there are many other false convictions that we have not discovered,” the Michigan study said. “In addition, a couple of strong demographic patterns appear to be reliable:

Black men accused of raping white women face a greater risk of false conviction than other rape defendants; and young suspects, those under 18, are at greater risk of false confession than other suspects.”

CapDefWeekly has more. Grits has a lot on snitching. Scott Greenfield feels vindicated. I will have more on this later.