Monday, June 12, 2017

Junk-science based false convictions in Houston lampooned by comedian

Someone has finally grokked and managed to convey in an accessible, understandable way the unmitigated travesty of justice surrounding drug convictions based on junk science in Houston. Comedian Samantha Bee aired a devastating 7-minute segment on false convictions in H-Town based on flawed drug field test results.

Grits has covered it. The Austin Statesman has covered it. Pro Publica and the New York Times Magazine have covered it. But no one has so succinctly captured the full scope of how these innocence claims play out in a corrupt system that railroads poor, innocent defendants better than this piece from Bee's show Full Frontal. Go figure.

This story has implications far beyond the borders of Harris County, or even Texas, as these field tests are used all over. And in some jurisdictions, if a defendant pleads guilty like one of the men in this story, crime labs don't always perform the followup test because of backlogs and cost constraints. So the most amazing part isn't the disparate treatment between the two defendants, but that these false convictions were ever brought to the light of day at all. Give it a watch, they did a great job with it:

Kudos to Inger Chandler from the Harris DA Conviction Integrity Unit and James Miller of the Houston Forensic Science Center (the bowtie was a nice touch) for their frank assessments of the problems with these commonly used instruments of junk science, which have caused an eye-popping number of false convictions. And that's just the ones we know about. They've only just begun to address the problem in Houston and no other jurisdiction in Texas is undergoing a similar review of cases based solely on field tests.

Four final notes: 1) It's a sad commentary on the state of Texas journalism that no outlet in this state ever told this story in such a compelling fashion, though all the pieces were there long before a comedian from Canada picked it up. Texas broadcast journos need to pick up their game. 2) The point about differences between outcomes for poor defendants and those with means plays out in every type of case, not just ones where the defendant was framed falsely convicted. 3) Shouldn't police departments just stop using field tests until the science behind them improves? And 4) the story speaks to the tremendous importance of the work at conviction integrity units. They only exist in a handful of DA offices, but they're exposing flaws in the system that we know also occur elsewhere. Like in this story.

Watch this excellent video clip and share it. Great stuff.

MORE: Two things: 1) it should be mentioned that, hours after I posted this, the governor signed HB 34 which includes a requirement that the Forensic Science Commission study the field test issue and report back to the Legislature by Dec. 1, 2018. Grits will try to cover this process, but I'd hope the MSM will, as well. 2) Nicolas Hughes, the lawyer at the Harris County Public Defender Office responsible for filing habeas writs in these cases, emailed these thoughts on the use of unreliable field tests:
Questions regarding the war on drugs aside, field testing presents a complicated problem. I’m all for the preliminary drug tests over the alternative – the officer’s hunch.

The question to me is what does the criminal justice system do with the information it gets from the preliminary drug test, knowing that the information is imperfect? Given that there’s a non-negligible chance that the test result will come back negative or at least differently than expected, I would hope that people would be released on PR bonds and/or a citation at least until a confirmatory test came back. I would also hope that officers receive some annual training on the tests and that agencies implement some quality assurance programs involving the tests.


Gadfly said...

Texas journos? By and large, "inside the Mopac" is a state-level version of "inside the Beltway."

Jefe said...

And how about Assistant Public Defender Nick Hughes, who filed all those successful writs?

Gritsforbreakfast said...

Indeed, didn't mean to leave out HCPDO efforts, Jefe - just mentioned Inger bc she was featured in the clip.

Anonymous said...

Why was a bill needed to get the Forensic Science Commission to tackle this? If the FSC was worth a crap, they would have initiated an investigation back in 2014 when it was noticed that the number of wrongful convictions sky-rocketed. As they did with their Willingham/Arson investigations and Bite Mark investigations, they basically have carte blanche to investigate anything forensic science related.

Texas Code of Criminal Procedure 38.01, Section 4 (a-1) "The commission may initiate for educational purposes an investigation of a forensic analysis without receiving a complaint..."

Not until February 2017 was this topic discussed at the TFSC meeting, and only in response to the Tim Cole Exoneration Review Commission's suggestion.

I guess if no one is watching the TFSC, they can sweep everything under the rug. Unreliable and Unaccountable.

Out of sight, out of mind, the wrongfully convicted will do the time.

Anonymous said...

By statute, the TFSC must evaluate complaints, and may undertake education research without receiving a complaint. The best way for issues to be brought to the Commission's attention is by filing a complaint, which would guarantee an evaluation (although not necessarily an investigation). Anyone can file a complaint.

In this instance, if no one was concerned enough about the issue to file a complaint with the TFSC, it is hard to see how the Commission can be faulted. Except gratuitously, of course.

Anonymous said...

Interestingly, if the District Attorneys Office of Houston had recognized the discrepancy between the field tests and the lab tests in 2014, why didn't they submit a complaint to the TFSC? Is it because they don't want to compromise their convictions? Why didn't the Defense Attorneys submit a complaint? Ignorance? Distrust for the TFSC's competence? Anyone can file a complaint. Yet...

If the Texas Legislature has to go so far as to create a bill to inform the TFSC as to what they should be doing and assign a due date for their homework, then clearly the TFSC is not paying attention to their forensic duties. And, their late-to-the-party discussions at the Meeting in February 2017 regarding the well-known piss poor drug field tests was initiated not because of any complaint, but because of the TCERC and the requested assistance by Representative Lozano for clarification of the tests. No formal complaint was submitted. There's no doubt, however, the TFSC would have dismissed any formal complaint given that the field test doesn't fall within their jurisdiction as defined by 38.01.

And, insofar as receiving complaints, by statue, the TFSC MUST investigate and they MUST create a report for the public. The words of 38.01 are very specific. Evaluation is not the same thing as investigation. In fact, the word "evaluation" only appears in 38.01 as a function of the investigational report. You have misinformed Grits' readers by suggesting that the TFSC optionally may investigate a complaint that falls within their purview.

Unfortunately, the equally misinformed general counsel of the TFSC also has decided that they don't have to investigate complaints (that fall within the definitions of 38.01). They simply rubber-stamp anything the crime lab states without actually verifying the truthfulness of the statements. Complaints of negligence or misconduct magically become "general allegations" and are dismissed without question. If a complaint can be loosely interpreted as "general allegation", then they can dismiss the complaint, no investigation, no report for the public. Reading comprehension apparently is not her forte.

Remember, Cecily Hamilton's 2010 complaints were deemed "general allegations" by the TFSC. But now, there is no APD DNA lab. See how that works?

Reading the minutes of the February meeting will give you stabbing pains in your head.

Gritsforbreakfast said...

@10:10 - So, "the field test doesn't fall within their jurisdiction" but they "MUST" investigate? Uh ... no. It's simply false that they are required to accept and fully investigate every complaint. The AG has said they have more leeway. They only "must" issue a report if they choose to undertake an optional investigation.

Also, the reporting requirement is for crime labs, not the DA, who in fact has sought to provide redress in this situation. Field tests are an unusual forensic tool because they're used by cops, not lab workers.

Should also be said that APD's DNA lab closed because of issues brought to the fore by the FSC. You can complain they should have acted earlier, and I agree dismissing Hamilton's complaint was a dark mark for the agency. OTOH, the lab wouldn't have closed without their work on DNA mixtures.

With all that said, it should be mentioned that, since I left IPOT, I don't think any advocates are working the FSC on an ongoing basis. Like most agencies, they respond to squeaky wheels (not anonymous whining but people willing to speak up under their own name).

Finally, there have been times when the FSC took crime labs word for it on issues and times they didn't. Usually it depends on what other evidence is available. Nobody's perfect, and they aren't either. But we know much more about forensic problems in Texas because they exist than we would if they didn't, which appears to be what you'd prefer.

Anonymous said...

Grits, you stated, "So, "the field test doesn't fall within their jurisdiction" but they "MUST" investigate? Uh ... no."

This is not what was stated in 10:10 above.

Have you read 38.01? The TFSC "MUST" investigate and write a report if the complaint falls within their jurisdiction (post-2005 events, and accredited crime lab, and forensic discipline listed in the statute). They don't get to pick and choose which complaints to investigate. If those three criteria are satisfied, they MUST investigate. They "MAY" investigate scientific problems if the forensic discipline falls outside of their jurisdiction (e.g. arson, bite mark analysis). The highly questionable field drug tests fall under the "MAY" category, thus the TFSC "MAY" investigate and "MAY" write a report. The AG opinion in 2011 gave the TFSC more leeway for the "MAY" investigations. HB34 took a "MAY" investigate forensic discipline (field drug testing) and changed it to a "MUST" investigate. Why did the Tx Leg feel the need to do this? Obviously, they don't trust the TFSC to perform an investigation that the TFSC could otherwise optionally disregard because the discipline doesn't fall within their jurisdiction as stated in 38.01.

The AG Opinion in 2016 (KP-0055) stated that the TFSC MUST create reports for the public in accordance with their statutory obligations (i.e. for those complaints that fall under the MUST investigate category.) Because the submitted complaints to the TFSC could contain detailed Brady Material information of negligence/misconduct which may be exculpatory in nature, public reports written by the TFSC fulfill potential Brady Material disclosures. Prosecutors and Defense Attorneys are more likely to get the relevant info if the TFSC provides a publicly written report detailing the events. Essentially, if the TFSC chooses to not investigate, and they don't create a report as they are required, they the could be effectively withholding Brady material contained within the details of complaint. If there is no TFSC report, Prosecutors and Defense Attorneys may not know that negligence/misconduct took place inside the lab. This is why the TFSC doesn't get to pick and choose which complaints to investigate that satisfy the three criteria within 38.01. Lawyers are depending on these reports.

1 of 2

Anonymous said...

2 of 2

"...there have been times when the FSC took crime labs word for it on issues and times they didn't. Usually it depends on what other evidence is available..."

Arguably if they investigated, they might find more evidence, no? By analogy, if someone (an anonymous whiner perhaps) called the fire department claiming there was an ongoing fire, wouldn't it be the responsibility of the fire department to find evidence of an ongoing fire rather than sitting on their duffs proclaiming categorically that there is no fire? If they don't ask questions, they won't get answers. If they don't show up at the scene, they won't find a fire. Complainants and subsequent Lab Responses are a "He said/She Said" scenario. The TFSC must validate/invalidate each side's story by investigating and asking questions and performing on-site audits. (Note to Grits: The TFSC Complaint Submission Form has a section for anonymous, complainants. You do not have to give your name to report a complaint. Anonymous complaints are no less credible than those with an attached name.)

Finally, no commentor has stated the the TFSC should be dissolved. The criticisms were that the TFSC wasn't doing their job according to statue, they weren't being thorough with their investigations (gullibly accepting lab statements as truth), and they weren't responding to contemporaneous forensic failures reported in the news.

"...Nobody's perfect, and they aren't either..."

Hence the need of the Tx Leg to create a bill to spoonfeed the TFSC. Frankly, it's embarrassing.

Anonymous said...

Need an example, Grits?

In 2009 a lab supervisor declared in open court that her lab analysts were using an expired chemical (used for identifying blood) over an extended period of time; "they were negligent" (her words). This was reported to the TFSC in 2013, but not as a self-disclosure. The lab supervisor's remarkable statements were reported to the TFSC by a witness to the proceedings.

Did the TFSC perform an investigation? Nope.
What was their excuse? They don't believe court testimony from lab analysts fall within their obligations for investigation.
What could have been discovered with a TFSC investigation? At least 5 different expired chemicals were being used, some as old as a decade. Sometimes, these chemicals were used in combination with one another. No validation experiments for the lab protocols existed. Also, analysts were creating reagents using chemicals that weren't part of the lab protocols, and wild deviations from lab protocols were not being documented by the lab analysts or lab management. And much, much more.
How many criminal cases could be affected? Thousands upon thousands, dating back to 1997.

Didn't hear about this, did ya?

That's because there was no report from the TFSC. The statements from the Lab Supervisor can be used to impeach her competency as a forensic scientist and any analyst that she trained or was under her supervision. But attorneys don't know this.

Anonymous said...

Need another example?


NO. 03-10-00665-CR
Laura Hall, Appellant
The State of Texas, Appellee

...Counsel went on to allege that the documents attached to the email he had received were not the "short synopsis of personnel problems" that prosecutors had disclosed to counsel prior to trial. Instead, counsel claimed, the documents contained allegations "regarding testing and quality assurance" and "management of the lab." Counsel included in his motion a summary of Hamilton's complaint against Carradine, which included allegations that Carradine had allowed a lab analyst with "very significant issues and problems" to perform independent DNA analysis and that Carradine had "misconstrued and misrepresented the facts about the quality assurance problems in the DNA unit to upper management."

Counsel concluded that, "[r]egardless of the truth of the matters asserted in these documents, it is clear that this was Brady material and any Assistant or Elected District Attorney had an immediate duty to bring the information to the attention of the Defense by providing all the documents in their possession." Hall makes the same argument on appeal: "Whether those allegations were true or not is beside the point. Substantive allegations had been made, and the allegations were known by the state's counsel to have been made, and the state was obligated to inform Ms. Hall of the allegations, if for no other reason than for impeachment purposes." (9)

Hamilton’s allegations, as far as the record reveals, implicated Carradine’s supervisory and management skills, not her abilities as an analyst. Also,there is no indication in the record as to the time frame of the allegations against Carradine and whether those allegations occurred during the time period in which Hall’s DNA was tested...

Yet, in 2016, the entire APD DNA lab is shut down for events that date back to 2009. Everyone in the lab was fired or reassigned to a non-forensics position because they were deemed "untrainable". Had the TFSC performed their duties in 2010 when the complaint was submitted, this could have been avoided. The defendant could have received a fair trial.

Gritsforbreakfast said...

Now, my anonymous friend. Give me similar examples from a state without a Forensic Science Commission?

Except, oh wait ... you can't ... because we'd never even know about the allegations in the first place.

Also, I've agreed with you repeatedly across many posts that the Hamilton investigation represents the agency's nadir and is its greatest dark mark, not sure why you insist on repeating yourself as though you've made a new argument.

Not that I can convince a zealot, but the statute says they shall investigate allegations IF they meet two criteria. They must involve an "allegation of professional negligence or professional misconduct" - lesser allegations or problems with underlying science, as with DNA mixtures, don't rise to that level unless there's a negligence or misconduct factor. Second, the misconduct or negligence must rise to such a level that it would "substantially affect the integrity of the results of a forensic analysis conducted by a crime laboratory." That's a judgement call that they were appointed to make and you were not, though you're obviously free to express your opinions about what they do.

IMO Lynn Garcia does as much as she's allowed to do and more than most would do, and unlike you she signs her name to every jot and tittle and accepts the professional consequences, (which include anonymous blog commenters trashing her online). The commissioners are all Republican appointees and mostly from forensic professions. I'm not sure if you expect them all to be hair-on-fire radicals, but if so, you should modify your expectations.

Anonymous said...

You only know about the allegations if the Forensic Science Commission makes them publicly available. You know nothing about the information they receive but choose not to publicly investigate. You don't know what you don't know. It's the same strategy used for decades by crime lab managers. If they don't acknowledge a problem in their labs through statement or documentation, attorneys don't know to ask about the problems in court. The assumption of "no news is good news" is in full affect.

see -- Tainting Evidence: Inside the Scandals at the FBI Crime Lab, by John F. Kelly and Phillip K. Wearne

Anonymous said...

Grits, are you stating that the APD's misunderstanding regarding incorrect analysis of DNA mixtures doesn't constitute negligence, and that this "lesser allegation" doesn't rise to a level as to "substantially affect the integrity of the results of a forensic analysis conducted by a crime laboratory" -- hence the TFSC is not required to investigate? What about allegations of DNA contamination that the lab analysts accidentally(?) failed to documented?

How about the lab analyst's use of expired reagents, also done at the APD lab? Is this negligence? What if a lab supervisor admits under oath that her analysts were negligent for using expired chemicals? Is this admitted negligence assigned to the analysts, or the supervisor, or neither? Maybe the lab supervisor's in-court declarations of negligence were incorrect (a fabrication), in which case her assertions constitute perjury. What are the definitions of the "negligence and misconduct factors" you speak of?

The two determinations (you mention above) are made post-investigation, not pre-investigation. The two determinations are not used to establish cause FOR performing an investigation; they are conclusions OF an investigation. TCCP 38.01 actually defines what an investigation entails. Per 38.01 the TFSC's written investigation reports for the public include, "Sec.4(b)(1)(B) whether negligence or misconduct occurred...;(b)(1)(D) observations of the commission regarding the integrity and reliability of the forensic analysis conducted." That is, the TFSC's determination of negligence/misconduct (yes/no) is to be stated in the written report, which is created AFTER each investigation. They also determine if the events tarnish the reliability of the results--again, determined AFTER the investigation. The TFSC "judgement calls" can only happen after all the information is scrutinized. Even if they find that there is no negligence/misconduct, an official report still has to be created stating how the Commission came to those conclusion. You can't put the cart before the horse. Attorneys must know the scientific reasoning for questionable lab reports that are presented in trial (or, most likely during plea bargains).

I would argue that the TFSC couldn't possibly determine pre-investigation if allegations truly involved negligence or misconduct at first reading of the complaint or the lab's first response (or self-disclosure). An investigation would have to occur. As a hypothetical, if a complainant alleges that lab analysts were not following lab protocols, and the lab's response was that the lab analysts WERE following lab this negligence? Could this deviation, if it happened, substantially affect the integrity of the results? Is the TFSC required to investigate? To answer the second question, the first question must be investigated (e.g. What is the foundation for the allegations? What are the lab protocols? Have the lab protocols been scientifically validated? Were the lab analysts properly trained and do they understand the scientific reasoning? Or conversely, were the lab analysts directed by their supervisor to disregard lab protocols thereby affecting the outcome? Was there adequate lab oversight? Is the lab management negligent for allowing the lab analysts to deviate from protocols? Was the lab management negligent for failing to implement protocol change to meet accreditation guidelines or "best practices"?)

I have a feeling that the root-cause of Ms. Garcia's misunderstanding stems from John Bradley's brief tenure. Short and embattled as it was, his interpretation of the TFSC's responsibilities stuck. Any excuse to not perform a public investigation (and minimize disruption of adjudicated criminal proceedings) is good enough.

Anonymous said...

Need another example?

Taken straight from the TFSC Meeting Minutes, February 2017...

"Bexar County Criminal Investigation Laboratory self-disclosed an incident in which a former analyst misidentified the caliber of a bullet recovered at autopsy and subsequently failed to exclude it in comparison as having been fired from a particular firearm. At the time of the non-conformance, the lab did not require independent verification of inconclusive results...."

(That is, for this incident, a false positive was the outcome. When, who, and how was the non-conformance detected? Does the former analyst agree with the allegations of "non-conformance"? Is it possible that the lab's assessment of "non-conformance" is incorrect, and the results are actually a true positive? What are the error rates for this discipline, by this analyst, at this lab, following their protocols at the time? Are accredited labs required to have (of this discipline) independent verification for all conclusions, including positive and negative associations? Is the lab's protocol "best practice" for the times?)

"The lab has since implemented independent verification for all conclusions. The lab reviewed reports for a 4-year period from November 2011 to January 2015."

(Why was November 2011 chosen as the start date for re-assessments? What about those analyses before 2011 that reported inconclusive results? What about those analyses after January 2015?)

"Of 576 reports reviewed, 78 had inconclusive results with no independent verifications..."

(Is 576 the total number of reports for this time frame, or were there additional reports selectively not included for review? Of the remaining 498 reports reviewed, how many false positives and false negatives were discovered? Of the false negatives, was the District Attorney notified so that a criminal proceeding could go forward? Of the false positives found, were Defense Attorneys and the wrongfully convicted contacted?)

"26 of those 78 cases were submitted for reanalysis based upon client requests..."

(Of the remaining 52 cases, why weren't they resubmitted for analysis? Were all the clients contacted, and did they all respond with either yes or no? Is it possible that other non-conformance issues could be found upon re-assessment of these 52 cases?

"Through the review process, the lab determined that the incident was isolated and limited to one analyst who is no longer with lab."

(How did the lab come to this determination? Were any of the other analysts questioned about being properly trained? Is it probable that "non-conformance" only occurs 1 out of 576 reports in this lab? Was the former analyst questioned about his/her conclusions of this particular case? Was the former analyst questioned about his/her training at the lab? Did the former analyst agree with the allegations from his/her former employer? Can a root-cause analysis be determined? Was the former analyst's incorrect conclusions accidental, or purposely made? What documents did the TFSC review regarding this self-disclosure? Did they get all the documents that were requested? And finally, Is this negligence? Could this non-conformance have compromised the integrity of the assessment?


But, we'll never know the answers because no investigation occurred and no written report was created. The meeting minutes are the totality of any written report.

Anonymous said...

Management decisions, including those labeled as “human resource” decisions, can have a tremendous impact on the laboratory’s overall transparency as a key player in the criminal justice system. For example, a critical component of every laboratory’s quality program is effective root cause analysis. The ability of the laboratory to conduct a fair and thorough root cause analysis in the wake of a non-conformance is essential to the integrity of the laboratory. When the laboratory issues a root cause analysis that inequitably attributes responsibility to one analyst while downplaying management’s contribution to the same incident, the resulting environment may be one in which analysts are hesitant to report mistakes. This dynamic can have a chilling effect on laboratory self-disclosure, which contradicts fundamental concepts in both the established accreditation standards under ISO-17025 and Article 38.01 of the Texas Code of Criminal Procedure.

Chris Halkides said...

I wrote an introduction to presumptive and confirmatory testing that touches upon some of the issues covered in this entry. I commend Harris County for going ahead with testing when it did. link

Chris Halkides said...

Let me say at the outset that I oppose field testing by untrained personnel. Some have pointed out that presumptive test kits are cheap and produce false positives. For reasons given in the link above, I would respond in all seriousness that these are features and not bugs of presumptive tests, as long as they are followed by confirmatory tests. IMO the answer is not to do away with presumptive testing; rather, the answer lies in limiting the scope of damage that can be done from a false positive from a presumptive test.

Anonymous said...

There are no established error rates for the field tests, in part because their accuracy varies so widely depending on who is using them and how. Data from the Florida Department of Law Enforcement lab system show that 21 percent of evidence that the police listed as methamphetamine after identifying it was not methamphetamine, and half of those false positives were not any kind of illegal drug at all. In one notable Florida episode, Hillsborough County sheriff’s deputies produced 15 false positives for methamphetamine in the first seven months of 2014. When we examined the department’s records, they showed that officers, faced with somewhat ambiguous directions on the pouches, had simply misunderstood which colors indicated a positive result.

The Justice Department issued guidelines in 2000 calling for test-kit packaging to carry warning labels, including “a statement that users of the kit should receive appropriate training in its use and should be taught that the reagents can give false-positive as well as false-negative results,” but when checked, three of the largest manufacturers — Lynn Peavey Company, the Safariland Group and Sirchie — had not printed such a warning on their tests.

Anonymous said...

As one of the above commenters above pointed out, anyone can submit a complaint. It's as simple as filling in a brief form obtainable from the Forensic Science Commission's webpage.
However, Texas Reps. Smithee, Moody, and Canales took the tedious kafkaesque approach and wrote the "complaint" into HB34. They could have taken the simple direct approach, but instead they took the overtly complex route.

This is the legislature way of stating, "TFSC -- you done f'd-up. We no longer believe that you can handle contemporaneous forensic issues on your own. We no longer trust your judgement. Now, you have a statute to abide."

This is a much bigger deal than what Grits and the MSM imply. The TFSC has been playing pattycake far too long and without meaningful oversight. Now everyone is watching.

Anonymous said...

HB34 is not a rebuke of the TFSC, but an expansion of its responsibilities into operations outside of accredited laboratories, which was its original sole focus. Under its original enabling statute, presumptive field tests performed by non-laboratory personnel were outside the Commission's scope. HB34 begins the process of changing that.

Anonymous said...


As has been stated many times, "The commission may initiate for educational purposes an investigation of a forensic analysis without receiving a complaint, submitted through the reporting system implemented under Subsection (a)(1), that contains an allegation of professional negligence or professional misconduct involving the forensic analysis conducted if the commission determines by a majority vote of a quorum of the members of the commission that an investigation of the forensic analysis would advance the integrity and reliability of forensic science in this state."

This is why the FSC, in part, investigated arson/fire science (a non-accredited field), bite marks (a non-accredited field), and digital evidence (which is specifically excluded as a "forensic analysis" and thus outside the scope of the FSC - but they did performed an investigation anyway.)

The fact that the FSC had not initiated any kind of investigation on its own regarding the drug field tests, much less acknowledged the issues, is the problem. This should have been addressed long ago by the FSC and without any kind of legislative dictate. Instead, the nuclear option was exercised.