Habeas Corpus Blog

Extending Martinez

By JK

Let’s turn the lights back on around here. . .

Recently, the Supreme Court granted cert. on a really interesting habeas issue:  whether Martinez extends to ineffective assistance of appellate counsel claims.  Case is called Davila v. Davis.  SCOTUSblog page here.

Let’s start with a Martinez refresher: in 2012, the Supreme Court held in Martinez v. Ryan that the ineffective assistance of post-conviction counsel can represent cause to overcome a procedural default on an ineffective assistance of trial counsel claim.  In other words, a petitioner can argue that his post-conviction counsel was ineffective for failing to raise or properly litigate a trial counsel ineffectiveness argument.  If he can establish that, then he can obtain federal review of the trial counsel ineffectiveness claim (I am going to use IAC from here on out, it’s a lot of words to keep spelling out).

That’s the law now for trial IAC claims.  But what about appellate IAC claims?  Does the reasoning of Martinez extend to those types of claims?

I am going to steal this clever framing from a colleague, but in some ways it boils down to this: It’s the Ninth Circuit vs. the rest of the country on this one.  The Ninth Circuit in a case called Nguyen v. Curry held that Martinez extends to these types of claims.  Every other court has said no. 

Based on the infamy of the Ninth Circuit, one may think that it is the kiss of death for the habeas petitioner to have the Ninth Circuit on his side.  But I wouldn’t jump to that conclusion so quick here.  The Ninth Circuit is not always wrong when it issues a pro-habeas petitioner interpretation on a procedural issue.*  That is not inconsistent with recent Supreme Court law, as the Court has actually been relatively generous in expanding habeas procedural issues under the Roberts Court (e.g. Martinez, Maples, McQuiggin). Ironically, Martinez was out of the Ninth Circuit and the Supreme Court reversed the Ninth Circuit for failing to rule in favor of the habeas petitioner.

*Of course, they do still get reversed on their pro-habeas petitioner procedural rulings as well.  This decision from last term is a good example.

Here’s why the petitioner could win.  The logic of Martinez supports its extension to IAC appellate claims looked at in broad terms.  One of the main motivating factors underlying Martinez was that the initial post-conviction review proceeding was the exclusive forum in which a petitioner was allowed to raise IAC trial claims.  A petitioner should be provided with an attorney to assist him with raising that type of constitutional claim.  Just like IAC trial claims, the initial review collateral proceeding is the first opportunity for a petitioner to raise an IAC appellate counsel claim.  Significantly, Justice Scalia in his dissent in Martinez acknowledged that the holding of Martinez would obviously apply to IAC appellate counsel claims.  He went so far as saying that the majority’s assurance that the holding was limited to the issues before it “insults the reader’s intelligence.”

But let me also advocate for the devil here.**  Is there something different between these two types of claims?  There is an argument to be made that there is.  As Justice Kennedy reasoned in Martinez, IAC trial claims often require investigation into matters outside the trial record.  Something that a habeas petitioner cannot do while incarcerated.  But appellate counsel claims don’t require that type of investigation.  Appellate issues are based exclusively on the lower court record.  Outside of having legal experience, wisdom, and skill, habeas petitioners do in theory have the issues at their disposal while incarcerated (assuming they have the trial record with them).

** No, I obviously do not believe the State is the devil.  It’s a play on words on the expression “playing devil’s advocate.” [Insert winky face emoticon here]

But I don’t find that reasoning particularly persuasive.  As I imply in that final sentence, habeas petitioners are not in a good position to raise these claims.  They don’t generally have the legal ability to identify appellate issues.  They are incarcerated.  They don’t have a lot of access to legal materials.  They don’t often have complete records from their trial.  Yet, criminal defendants have the constitutional right to the effective assistance of appellate counsel (yes, spelled it out there, but it seemed appropriate).  These claims can only be brought in the initial review collateral proceeding.  Representation by counsel in those proceedings to raise a violation of a guaranteed constitutional right seems pretty vital to me.

Martinez was 7 to 2.  If the court remains at 8, it will take a shift of four members of the majority for the Court to reject the Martinez extension.  But I guess it is more complicated than that.  If the habeas petitioner loses three, then it is 4 to 4.  The individual petitioner will lose in that situation, but at least petitioners in the Ninth Circuit would still be happy.  And the issue would remain undecided.  

If the Court gets back up to 9, the numbers game would change a little.  Assuming the new judge votes like Scalia, that will add another judge to the anti-Martinez extension crew.  In that situation, a three-judge shift will mean a lose for all habeas petitioners.

So it really boils down to whether the habeas petitioner will lose all of Roberts, Kennedy, and Alito.  Only needs to hold on to one for the win here.*

* Yes, I am assuming Ginsburg, Kagan, Breyer, and Sotomayor are reliable votes here. 

If you want to read more about this, I enjoyed this write up about the case.

SDNY Grants 2254 Resurrected by Martinez v. Ryan

by Matthew Keller

Court finds state trial counsel ineffective for failing to spot "straightforward" Massiah issue during trial.

Smith v. Fischer, No. 07 Civ. 2966 (S.D.N.Y. Jul. 16, 2013)

        In 2012, the Supreme Court held in Martinez v. Ryan that having an ineffective lawyer (or having no lawyer at all) in a state collateral proceeding could excuse a petitioner's failure to raise an ineffective-assistance-of-trial-counsel claim.  I thought this was a good decision because it would allow the lower federal courts to expend their scarce resources evaluating the merits of ineffectiveness claims rather than hashing out comparatively esoteric considerations of cause and prejudice.  In light of the onerous AEDPA standard, I didn't expect too many petitions to be granted after a merits review.  But last week, the Southern District of New York did just that.  Smith v. Fischer may very well be the first decision in the Second Circuit (or anywhere?) granting a 2254 habeas petition that had been resurrected by Martinez. 

        In March 2012, the district court denied Patrick Smith's 2254 petition as procedurally barred: the state court decision was based on Smith's failure to raise his ineffectiveness-of-trial-counsel claim in his first 440.10 (NY state collateral) motion.  Smith had filed that first collateral motion pro se. 

        A few weeks later, SCOTUS issued Martinez.  Two days after that, Smith moved for reconsideration.  It took the district court eight months, but this past January it decided that Martinez applied and, therefore, the court could reach the merits of Smith's petition.

The Case of the Serendipitous Snitch

        In 2003, a jury convicted Smith of intentional and felony murder for the killing of a security guard during an armed robbery.  Smith was sentenced to twenty to life.  He exhausted his direct appeals in June 2006 and filed his first 2254 petition in 2007. 

        The prosecution had faced substantial obstacles to trying Smith.  He was not arrested until seven years after the crime.  Indeed, the defense was able to have four counts of the six-count indictment dismissed on statute of limitations grounds.  The prosecution had eyewitness ID testimony, but that testimony  described someone who didn't resemble Smith very much and, in any event, those IDs were seven years old.  This gave the defense ample opportunity to impeach the State's identification witnesses based on the passage of time and degradation of memory.  Finally, during pretrial motion practice the trial court excluded Smith's incriminating statements to the police, further weakening the State's case.

        Then, a few weeks before trial, the prosecutor annouced a surprise witness - a jailhouse snitch to whom Smith allegedly confessed.  At trial, William Ferguson testified that, over the course of several days while they were incarcerated together, Smith told Ferguson about his involvement in the crime.  On cross, Ferguson admitted that he pressed Smith for information and attempted to gain Smith's confidence by "sharing his own background" and discussing his own crime.  On postconviction review, Smith's ineffective assistance claim was that trial counsel should have moved, pursuant to Massiah,* to exclude Ferguson's testimony because Ferguson was acting as the government's agent when he elicited the incriminating statements.

*  In Massiah v. United States (1964), the Supreme Court held that  the Sixth Amendment prohibits the government from eliciting statements from a defendant after the right to counsel has attached.  In United States v. Henry (1980), the Court held that an inmate who informs the government of a defendant's incriminating statements becomes the government's "agent" with  respect to later-elicited statements.

The 2254 Decision

        The Southern District reviewed Smith's ineffectiveness claim de novo.  It held that the state court did not make a merits determination of the ineffectiveness claim, even though the state court had referred to the claim as "meritless."  Prior to that, the state judge stated that it "declines to reach the merits of Defendant's [ineffective assistance of] counsel claims." The district court took the state court at its word. It held that the state court's "bare references"* to the merits of the claim following its assertion of a state procedural bar did not "undermine state court's ruling that it 'declines to reach the merits of case.'"

*    This appears to be a dig at the state court decision, which analyzed the "merits" of Smith's ineffectiveness claim in approximately 12 words: Smith's "bare claims of ineffective assistance do not meet the Strickland" standard.  

        The district court had little problem finding ineffectiveness on the part of trial counsel.  It held that "[t]he Massiah issue in this case is straightforward:"

"Mr. Ferguson testified at trial that [he] had agreed to testify for the prosecution before deliberately eliciting incriminating statements from Smith.  Mr. Ferguson's testimony on this issue was enough that defense counsel knew, or should have been aware, that a Massiah violation had occurred and that the statements should be suppressed."
 

        As for Strickland prejudice, the district court had this to say about the probability of an acquittal in the absence of Ferguson's testimony:

"due to the weaknesses the eyewitness testimony, which the trial judge noted, and the lack any scientific or physical evidence linking Mr. Smith to the crime, Petitioner has shown a 'reasonable probability' that he would have been acquitted but for his counsel's failure to suppress Mr. Ferguson's testimony under Massih.  Aside from Mr. Ferguson's testimony as to Mr. Smith's confessions, the evidence at trial consisted of eyewitnesses whose descriptions of the perpetrators were inconsistent with that of Mr. Smith, and whose memories had faded over the seven years between the and the trial. Mr. Smith was not arrested until six after the crime, and there was no physical or scientific evidence tying him to it. Given the limited evidence aside from the informant's testimony, the prosecutor emphasized Mr. Ferguson's statements in summation, including those statements which Mr. Ferguson elicited after he had agreed to testify. During deliberations, the jury asked for read backs of Mr. Ferguson's testimony . . . and returned a guilty verdict the following day. The state court judge who tried the case observed, '[i]t is reasonably clear to me, based upon the quality of the eyewitness testimony in this case, that the jailhouse informant had to be a reasonably important witness.'"  

         Perhaps anticipating that the Second Circuit might disagree that the state court decision was not on the merits, the district court held that the state court denial also didn't survive AEDPA deference: the state court's application of Strickland was "objectively unreasonable" because, "[g]iven the similar critical importance of Mr. Ferguson's testimony to the central issue before the jury in this case, no court could justifiably find that there was not a 'reasonable probability'  that, but for defense counsel's failure to suppress Mr. Ferguson's testimony, Mr. Smith might have acquitted."

        Finally, kudos to SDNY District Judge Sweet for calling it like he saw it here, once again demonstrating the benefits of life tenure free from electoral and other political pressures.  On the issue of whether the police had sent Ferguson in to get something from Smith, the judge was skeptical that it could be otherwise:  

"That Ferguson appeared as an informant on the heels of the trial judge's preclusion of Smith's statement to the police, that the police officers took no notes of their meeting with Ferguson or Ferguson's initial call to [an NYPD Detective], and that Ferguson, as an inmate, was able to have Smith brought to him in the library by a corrections officer all raise serious questions as to whether the police engaged in an intentional effort to secure Smith's incriminating statements."

        That's not Monday morning quarterbacking (which, to be fair, federal courts do indulge in from time to time when reviewing state court trial proceedings).  That's calling the state court out for a serious lapse in sound trial management.  That's what 2254 should be about.

Second Circuit Grants Habeas Claim for Ineffective Assistance

Your attorney doesn’t have to be found guilty of murder to be ineffective, but it helps.

by Matthew Keller

     It is a fundamental tenet of criminal defense that your client’s unrelated bad acts cannot be used against him.  To do so would unfairly tempt a jury to find guilt not on the charges, but on the “bad man” theory that, if someone did wrong in the past, it’s more likely that they did wrong in the case at hand.

   Do lawyers deserve the same treatment?  According to the Second Circuit, apparently not.  Last week, in Gonzalez v. United States, 10-3630-pr (2d Cir. July 10, 2013), the Court remanded Pedro Gonzalez’s case for resentencing for the second time.  The only difference between the first and second remands: in the interim, the attorney was convicted of murdering his wife. 

The $35,000 No-Frills Retainer

     Pedro Gonzalez owned a small contracting company in upstate New York.  However, it appeared he had become involved in a drug trafficking operation that used drugs and cash to bribe a corrupt immigration official in exchange for green cards for illegal immigrants.  In 2001, he pled guilty to eight narcotics and bribery charges for his part in the scheme.  At his plea hearing, Gonzales admitted that he had made several deliveries of cocaine or heroin and that he had “produced over two kilograms of heroin throughout the course of this conspiracy.”  Consequently, a judge ordered him imprisoned for seventeen and a half years. 

     At his sentencing, Gonzales asked for his plea back.  He told the court that his attorney, Carlos Perez Olivo, had not sufficiently explained to him the charges or the plea process.  Gonzales claimed that Olivo, after accepting a $35,000 retainer, told Gonzales to sign the Plea Agreement without allowing Olivo to read it or even explaining to him what it was.  Olivo then told Gonzales to “‘agree with everything that was being said’ at the plea hearing in order ‘not to upset the judge’ or risk losing benefits the government had promised.”  In addition, Olivo did not submit a sentencing memorandum and did not meet with Gonzales prior to sentencing.  During the sentencing Olivo spoke only two sentences in favor of a sentence at the low end of the Guidelines range. 

      In a follow-up letter to the Court, Gonzales also offered what sounded like a defense to the charges or, at least, a reasonable mitigation argument. He explained that he became involved with the conspiracy after one of his employees introduced him to “a friend” who offered Gonzales legal services in order to obtain green cards for his undocumented workers.  The friend assured Gonzalez that it was all legal.

     Not quite.  Gonzalez went to a meeting at an Albany restaurant, he said, for the purpose of learning more about what the immigration lawyer could offer.  Instead he met a group of people who thereafter “threatened and used” him in order to compel his participation in their narcotics activities.  Gonzales said that a Government informant “called me and told me if I did not get the drugs for him, they were going to kill me.”  This is a compelling narrative: the bad guys, through threats and coercion, strong-arm a hardworking small businessman into doing their dirty work.  If Gonzales could support his version of events, he had a strong defense of duress.*

   *The Sentencing Guidelines also recognize a coercion basis for a downward departure where “the defendant committed the offense because of serious coercion, blackmail or duress, under circumstances not amounting to a complete defense.” See USSG § 5K2.12.)

First Habeas Motion

     Soon after he was sentenced, Gonzales filed an initial § 2255 motion claiming ineffective assistance of counsel based on Olivo’s failure to appeal the court’s denial of the motion to withdraw the guilty plea.  The district court denied the motion on grounds that Gonzalez’s plea agreement contained a valid appeal waiver.  Gonzalez then sought, and the Second Circuit granted, a COA on the issue of whether Gonzalez had asked Olivo to file an appeal – the Circuit remanded for factfinding on that issue.  On remand, the government submitted an affidavit from Olivo, “stating at length that he had not been paid and, far more briefly, that Gonzalez had told Olivo that Gonzalez did not wish to appeal.”  Apparently, Olivo believed that no respectable lawyer could be expected to pursue a defense for just $35,000.

     In February 2006, over four years after he was sentenced and his motion to withdraw his plea was denied, the district court credited Olivo’s affidavit and again denied Gonzalez's § 2255 motion.  Gonzalez again sought a COA, and the Second Circuit again granted it to review “whether the district court properly decided the factual question – i.e., whether Gonzalez asked Olivo to file a notice of appeal – solely on the basis of affidavits.”

     While that appeal was pending, things started to fall apart for attorney Olivo.  In August 2006, he was disbarred in New York State for, among other things, defaulting on a former client’s appeal and, to avoid the consequences of that default, misrepresenting to the client that the appeal was meritless.  (It also turned out that Olivo previously had forfeited a license to practice law in Puerto Rico due to similar charges of malfeasance). Three months later, Olivo was shot and his wife was tragically murdered.  Olivo told police that the two of them were driving to their home in Chappaqua from New York City when another car ran them off the road and shot at them.  His wife was struck in the back of the head; authorities described Olivo’s wounds as “minor.”

     Maybe it was the disparity in the alleged assassin's accuracy that caused authorities to focus their investigation on Olivo.  Whatever the reason, he didn’t help his case by holding a creepy yard sale in which he sold his dead wife’s boots, pajamas and other personal items.  He was charged, and later convicted, of the murder.

     While Olivo’s murder case was proceeding, in February 2007 the district court granted Gonzalez’s § 2255 motion and vacated his sentence.  However, the court resentenced him to the same terms as the initial sentence.  Gonzalez again appealed, asking (again) for his plea back and, barring that, for a new sentence based on Olivo’s ineffective assistance.  The Second Circuit affirmed in November 2008, but granted leave for Gonzalez to file a second § 2255 motion “[i]n view of the existence of facts, not previously available, that might relate to the performance of counsel (including those that might bear on counsel’s effectiveness at the time of Gonzalez's guilty plea).” 

The Second § 2255 Motion

     In his second habeas motion, Gonzalez “drew the court's attention to the new fact that in 2008, Perez-Olivo was convicted of murdering his own wife and was currently serving a prison term of 25 years to life.”  Despite that new information, and despite a 63-paragraph sworn affirmation “describing in detail the events leading to his arrest, beginning with his being contacted by one of the government's confidential informants,” the district court denied the motion.  The district court implicitly conceded that Olivo had been Strickland deficient with respect to both the plea and sentencing but, it held, there was no prejudice: the court concluded that Gonzalez “fail[ed] to show that the deficient representation he received from trial counsel adversely affected the length of his sentence.”

     Last week, the Second Circuit reversed.  With respect to Gonzalez’s plea, the Court arguably created* a new rule in its Strickland jurisprudence: “the fact that an attempt was made to withdraw the guilty plea and go to trial may not be dispositive on the issue of IAC prejudice; however, it is a factor that must be considered by the court in assessing whether there is a reasonable probability that but for substandard performance by counsel, the defendant would have chosen to eschew the plea and go to trial.”  With characteristic pithiness, the Circuit reasoned that in determining “whether there was a reasonable probability that, absent [counsel’s] allegedly substandard performance, Gonzalez would have gone to trial, the district court did not mention the plainly relevant fact . . . that Gonzalez had actually attempted to . . . go to trial.”   

     *  I say that this is “arguably” a new rule because it is just as arguably dicta.  The court held that the record was devoid of evidence of deficient conduct.  In essence, the Court said that, despite having twelve years to develop his duress claims, Gonzalez had not presented any proof that “Olivo failed to investigate any aspects of the case that could have led to a successful defense by Gonzalez if the charges against him were tried.” 

 

     Gonzales fared better, however, with his ineffectiveness claims regarding sentencing.  The Circuit reversed the district court’s finding of no Strickland prejudice:

“[t]he hypothetical sentencing arguments proffered by Gonzalez's new counsel, along with the district court's reference to some of them as potentially ‘effective’ arguments, suffice to undermine our confidence in the outcome of Gonzalez's original sentencing and thus to show that the reasonable-probability standard with respect to the sentencing claim was met.”

 

     What role, ultimately, did Olivo’s murder conviction play in the Second Circuit’s decision?  Your answer to that question probably says something about whether you are at heart a legal realist or a formalist. But ask yourself: if there was no evidence to support a duress defense, what can Gonzalez possibly have used to justify a departure on that basis?  The Guidelines provision provides for a downward departure where the evidence does not support a "complete" duress defense, but Gonzalez still needed some hard evidence to support his claim that the other conspirators threatened him.  If you accept the Court's conclusion that Gonzalez again failed to come forward with such evidence, the only difference between this remand and the previous one is Olivo's murder conviction.   

     Logically, Olivo’s murder conviction - and his deficient performance in other cases - do not prove that he was ineffective in Gonzalez’s case.  But the Second Circuit’s two remands of this case for a second and third look suggest that the Court is (wisely, I think) not prepared to trust Gonzalez’s fate to a disbarred murderer.

Judge Garaufis (EDNY) Takes NY Trial Judge to Task; Prevents State From Retrying Old Murder Case

By Matthew Keller

 

Lopez v. Miller, No. 02 Civ. 3988
(E.D.N.Y. M&O filed Jan 16, 2013)

        When I read this case out of the federal court in Brooklyn back in January, I was excited to post about it.  But eventually so much time passed that I was afraid any post would be time-barred (like AEDPA, HCB's head honcho JK sets a high standard for timely blogging).  But the Supreme Court's decision in McQuiggin v. Perkins a couple of weeks back opened a door for me.  That case confirmed that there is an actual innocence exception to AEDPA's one-year limitations period.  However, Lopez v. Miller applied this exception back in January . . . and went on to grant a 2254 petition based on ineffective assistance of trial counsel.  A couple of months later, the court took the extraordinary step of forbidding New York from retrying the defendant, and ordered his release.  In response to Justice Scalia's anti-habeas diatribe in the McQuiggin dissent (discussed in JK's recent post on McQuiggin), I think it's appropriate to illustrate the type of case that McQuiggin now allows federal courts around the nation to address.

        Judge Garaufis’ opinion in Lopez is interesting for several reasons.  As I've already said, it foreshadowed McQuiggin by applying an actual innocence exception to an untimely habeas petition.  In doing so the court was following the Second Circuit's ruling last year, in Rivas v. Fischer (see HCB post here), that such an exception was required by due process.  Lopez is also rare in that the court grounded its habeas grant on a § 2254(d)(2) finding that the state court decision was based on unreasonably determined facts, as opposed to an unreasonable application of Constitutional law.

        Lopez is also noteworthy for reasons outside of habeas corpus law.  First, it has federalism implications; rarely does a federal judge take a state’s criminal justice system (and, especially, a state judge) to task as roundly as Judge Garaufis did here.  Garaufis wrote that the state court judge’s unexplained refusal to hold an evidentiary hearing to explore new evidence of innocence was “baffling” and resulted in a loss of important evidence (by the time the federal court hearing, two alibi witnesses were unavailable).    
 

        Second, Lopez is a stark example of justice delayed being justice denied.  Mr. Lopez filed his original § 2254 petition pro se in 2002.  Eleven years later a habeas court ruled that his state court murder conviction must be vacated because it resulted from a self-evidently bad witness ID, a lying prosecutor, and a baffling indifference to both of these by the state court judge. 

 
A “Rotten” Case 

        “The case of William Lopez began twenty-three years ago. It was rotten from day one.”  So begins Judge Garaufis’ 85-page M&O describing Lopez’s conviction for the 1989  murder of a Brooklyn drug dealer.  Pursuant to that conviction Lopez spent 23 years in prison.  Garaufis is unusually free with criticism of how New York state's criminal justice machinery operated in this case:  at one point he notes that, “the prosecution’s evidence was flimsy to begin with and has since been reduced to rubble by facts arising after trial.”  But he saves the real fireworks for the conclusion:
 

“Lopez’s habeas case presents a number of close legal issues, both because of the stringent standards governing federal habeas review and because of the unavailability of many key players. But what is far from close in the court’s view is that Lopez has been wronged by the State of New York. This wrongdoing has ranged from an overzealous and deceitful trial prosecutor; to a series of indolent and ill-prepared defense attorneys; to a bewildering jury verdict; and to the incomprehensible [New York Supreme Court] Justice Demarest, who so regrettably failed time and time again to give meaningful consideration to the host of powerful arguments Lopez presented to her. The result is that a likely innocent man has been in prison for over twenty-three years. He should be released with the State’s apology.”
 

The State should apologize?!?  I'm no historian, but I can't help thinking that it was humbling commands like this from Supreme authorities (i.e., the Crown) to the States that led in part to the American revolution.  Maybe that's why, even today, such direct criticism of state government by federal authorities is still rare.  And criticism in defense of a man convicted of murder?  Even when a man has been exonerated, many states in our nation treat him with slightly-concealed disdain (by, for instance, passing laws denying compensation to an exoneree who “contributed” to their wrongful conviction via a coerced confession or a guilty plea).*  What could have Judge Garaufis so exercised that he would suggest that a State actually apologize? 

            *  Such laws are an affront to our country’s tradition of liberty.  But I digress.  

Justice Delayed . . . A Really Long Time

       Lopez has a long procedural history that is worth summarizing.  Elvirn Surria dealt drugs in Brooklyn.  Late one night in 1989, two men shot and killed him in a Coney Island crackhouse.  A jury convicted Lopez of murder based entirely on the testimony of two witnesses, only one of whom identified Lopez in court.  That witness, Janet Chapman, testified that she saw the shooting through a half-open doorway.  She also testified that she had smoked ten to twelve vials of crack in the two hours prior to the shooting. The second witness came face-to-face with the shooter, and spoke with him, but could not recognize Lopez when she saw him in the courtroom.  During the trial New York Justice Carolyn Demarest stated that it was “not possible this defendant could have committed the crime” based solely upon the second witness’ testimony.

        Fast forward ten years.  Lopez’s state court appeals have all been denied.  In July 2002, he filed a pro se § 2254 Petition.  Almost two years later, in April 2004, the district court dismissed the petition as time-barred, since it was filed more than a year after the 1996 passage of AEDPA.  (Query number 1: Even granting the huge backlog of habeas cases in the Eastern District of New York, how on Earth does it take a federal court two years to determine that 2002 comes after 1997?) 

        Lopez moved for reconsideration of the dismissal, arguing actual innocence.  Shortly after his trial, Chapman told Lopez’s brother that she had falsely ID’d Lopez because of pressure from the police.  Ultimately, she provided an affidavit recanting her ID and stating that she testified falsely based on the trial prosecutor's promise that she’d get out of jail on drug charges if she testified.  And the kicker: Chapman also stated that “the district attorney told me never to tell anyone that we cut a deal about my testimony in exchange for my freedom.” 

        Judge Garaufis was concerned enough about these allegations that he granted the motion to reconsider and, in 2005, appointed a lawyer to investigate.  Fast forward another four years.  During that time counsel learned, among other things, that Chapman had died (Query number 2: why did that take four years to figure out?)  In 2009, Lopez moved to stay and abey his habeas petition so that he could exhaust state remedies.  A year later, Justice Demarest denied Lopez’s motion to vacate his conviction on the grounds that there was “strong evidence of guilt” at Lopez’s trial.

        One year after that, in November 2011, Lopez filed an amended 2254 petition.  In July 2012, Judge Garaufis ordered an evidentiary hearing on Lopez’s gateway actual innocence claim and his 2254 claim of ineffective assistance.  The court heard testimony from Lopez, his brother, and the trial prosecutor, Tess Allen.  In January 2013, the court reopened the hearing to allow new evidence from Cesar Diaz, who appeared via video conference from the Dominican Republic. Diaz, who was present at the scene of the murder, was shown a mug shot of Lopez taken soon after the crime and testified that he was “certain” that the person in the picture was not the shooter.

        Based on the new evidence presented at the habeas hearing, the court found Lopez’s gateway actual innocence claim to be credible (that is, “supported with new reliable evidence . . . not presented at trial”) and compelling (in that it warranted a conclusion that “more likely than not any reasonable juror would have reasonable doubt” if it had heard the new evidence), as required by the Second Circuit’s decision in Rivas.  In fact, the court stated that the new evidence was largely beside the point: “Even without the benefit of Lopez’s new evidence, a reasonable juror would likely have serious questions about Lopez’s guilt.”  The court called the case as presented to the jury a “toss-up at best” and took another swipe at Justice Demarest’s conclusion (in denying state post-judgment relief) that there was “strong evidence of guilt.”  That finding, the district court said, “contains no explanation and is, frankly, outlandish given the circumstances described above. The court gives it no deference.” 

Through the Gateway. . . .

        After the court admitted Lopez through the actual innocence gateway, it sustained his claim of ineffective assistance on the infrequently-used 2254(d)(2) ground that Justice Demarest’s decision denying Lopez’s IAC claim was based on an unreasonable determination of the facts in light of the evidence presented in the State court proceeding. Essentially, the court found that Justice Demarest could not effectively evaluate the claim because she had refused to hold a hearing to resolve a credibility dispute between Lopez’s trial lawyer, who said he interviewed two potential alibi witnesses, and the witnesses themselves, who said he did not.  The failure of the state court to resolve this credibility dispute via a hearing led to a “materially incomplete” state court analysis and, as a result, the state court decision was unreasonable. 

        The court then moved to the merits of Lopez’s IAC claim and found that trial counsel had been ineffective in failing to call either of two potential alibi witnesses.  Judge Garaufis ordered New York to release Lopez within sixty days “unless [it] has, by that point, taken concrete and substantial steps expeditiously to retry [him].”

        Following the decision, New York re-arraigned Lopez on murder charges, but after that failed to "offer[ ] any other proof that it intends to promptly retry him."  Accordingly, on March 20, 2013, Garaufis ordered Lopez to be "immediately and unconditionally released" from custody.  He also ordered that "the State is BARRED from retrying him for the August 31, 1989, murder of Elvirn Surria" and "the State shall EXPUNGE Lopez’s conviction from its records and all references to him in the public record."  The court stated it was taking the rare step of forbidding a retrial because it was "convinced that this is one of the extraordinary cases that warrants an unconditional discharge with no future threat of retrial."  In justifying its order, the court took a final shot at New York's handling of the case:  "due to the errors of a State court judge, a State prosecutor, and counsel, Lopez was subjected to an unconstitutional deprivation of liberty for twenty-three years."

        I have to say that Lopez did not have a slam-dunk IAC claim.  Judge Garaufis seemed to recognize this - he found that trial counsel was not deficient for failing to interview the potential alibi witnesses (there wasn't sufficient evidence to figure out whether he had) but rather for failing to call them at trial.  But counsel had explained to Justice Demarest at the trial that he had considered calling them but had decided that the jury would likely not find them credible.  In my experience, federal habeas courts almost always defer to such strategic decisions by trial counsel.  In any event, I think its safe to say that this IAC claim could have gone either way.  What really drove the district court, I think, was the state court's repeated failures to examine the case as the evidence of a wrongful conviction mounted (especially the evidence that the prosecutor had deceived the trial court about its deal with Chapman).  Here, the district court was able to justify the habeas grant on an iffy IAC claim in light of several breakdowns during trial.  So we will have to keep waiting for the case where a defendant with a credible and compelling claim of innocence does not have a colorable claim of a Constitutional violation. 

Postscript

        As far as I can tell, New York State has not yet apologized to Lopez.  And as the history of this case shows, the State did not do much to help him, even after it became clear that there were serious problems with his conviction.  But karma finds a way: New York ultimately did play a role in helping to free Lopez.  Another exoneree, Jeffrey Deskovic (released in 2006 after serving 16 years for rape and murder based on a coerced confession), used some of his $8M settlement with the State to start a foundation to assist other inmates with credible innocence claims.  Deskovic's foundation partly funded Lopez's habeas case.  

Martinez v. Pinholster: Coming to SCOTUS Soon?

By Alexandr Satanovsky

When a Strickland-deficient state postconviction lawyer fails to raise a substantial ineffective assistance of (trial) counsel ("IATC") claim, the federal courts will excuse the default in habeas. But what if the state lawyer does fairly present the IATC claim, but loses after shamelessly failing to develop the factual basis? This is the "Martinez v. Pinholster" tension, which JK discussed a few months ago, now percolating in some federal courts of appeal.

The eventual Supreme Court resolution of this conflict seems pretty inevitable, and some tea leaves may appear very soon: This very question* is presented in Gallow v. Cooper, a noncapital, state-on-bottom cert. petition out of the Fifth Circuit.  It is noteworthy because it is now scheduled for its fifth cert. conference, which is very rare and typically indicative of something brewing.  

* Gallow presents the following question: "Can a federal court consider new evidence to support a state prisoner’s application for habeas relief under 28 U.S.C. §2254(d), when the state court record was not developed as a result of incompetent and likely conflicted post conviction counsel[?]"

Curiously, neither the parties' briefing nor the opinion below engage the biggest knot in the Martinez-Pinholster dispute: whether the Supreme Court has the authority to carve out common-law equitable exceptions to statutes, and if so, whether it can recognize an exception that was not in the common law at the time the statute was enacted, and whether it can be predicated merely on an attorney's deficient performance, as opposed to a traditional and more compelling basis like miscarriage of justice.

The difference is with the nature of the limitation that the prisoner seeks to overcome: procedural on one hand, and the §2254(d) inquiry on the other. The former is a court-made common law doctrine intended to promote comity and respect valid state procedural rules. The latter is a statutory command, which significantly limits the Supreme Court's authority to craft exceptions. The closest authority discussing Court-made exceptions to a statutory command are Holland v. Florida and McQuiggin v. Perkins, both concerning the Congress-enacted statute of limitations.  In each case, the Court justified its exception (equittible tolling for extraordinary circumstances, and actual innocence, respectively) by noting that it was established in the judicial common law against which Congress drafted the statute; if Congress sought to abrogate the equitable exception, it would have been more explicit. In this case, the problem is different: the Martinez exception neither implicates a miscarriage of justice nor was established at the time Congress drafted §2254(d). 

I don't think anyone would disagree that, as a matter of policy and equity, a petitioner whose deficient state collateral lawyer failed to raise an IATC claim should be treated differently from one whose lawyer raised (and lost) an inexcusably maldeveloped claim. Both are victims of their respective state lawyers' incompetence, but for the former Martinez will provide the windfall of de novo federal review, while for the latter, because of a state court ruling on the merits, there is neither de novo federal review nor even deferential review aided by more facts. 

Perhaps there is a better way for the Supreme Court to resolve this inequity without mechanically recognizing an "exception" per Martinez. Martinez works against a judicially-crafted bar (procedural default), but if recent cases are any indication, it will not fare so well against a statutory limit (§2254(d), as interpreted by Pinholster). 

To harmonize the equities of Pinholster and Martinez, the Court could provide more clarification to the amorphous language in §2254(d)(1)-(2): What exactly is a "claim adjudicated on the merits"?  What can make for an "unreasonable" determination of the facts, or an "unreasonable" application of Strickland? For example, the Court could hold that it is "ureasonable" for a state court to evalute the the merits of a fact-intensive IATC claim on poorly developed pleadings. Alternatively, as alluded to in Pinholster, the Court could clarify what constitutes a "claim" under §2254(d), and hold that whenever the factual grounds in a federal pleading are different enough to create a substantial chance of being outcome-determinative compared to facts pled in state court, then the federal pleading is a new claim, to be analyzed under Martinez. Ultimately, even if the equitable principles embodied in Martinez may not justify a court-decreed exception to a clear statutory provision, they are quite persuasive for a court's interpretion of an ambiguous statutory provision.

Or, of course, there is one solution that is the simplest of all: the Court could revisit what it chose to punt in Martinez, and hold that prisoners have a federal constitutional right to counsel in all initial-review proceedings. This approach, plainly it seems to me, would compel that §2254(d) limits be bypassed upon a showing that a constitutional violation tainted the state court's adjudication.

*    *    *

So what is the deal with Gallow?  My prediction is that the Court will deny certiorari, possibly with dissenting and concurring opinions. This is a prisoner's cert. petition, and there is absolutely no chance of a summary reversal that would necessarily alter the Pinholster'd §2254(d) review. The Martinez-Pinholster resolution will come eventually, likely after more courts of appeals flesh out these difficult issues, which the Fifth Circuit surely did not in Gallow.  

Habeas Corpus - Martinez vs. Pinholster - Who wins?

By JK

Has the Ninth Circuit found another way around Pinholster?  Can a petitioner rely upon a Martinez argument in order to have a federal court review "new" evidence?  Boy, these pathways around Pinholster really are a hot topic in these parts.  

On Monday March 4, Judge Tallman, joined by many, if not all, of the other conservative judges on the Ninth Circuit, issued a published order dissenting from the denial of rehearing en banc in a capital habeas case called Schad v. Ryan.  

In an unpublished order, the three-judge panel in Schad remanded the case to the district court for that court to determine whether petitioner could rely upon Martinez to establish cause for a procedurally defaulted IAC claim.  Petitioner had raised an IAC claim based on counsel's failure to present mitigating evidence during the penalty phase.  In his habeas petition, petitioner was relying upon mitigating evidence that post-conviction counsel ("PCC") failed to present to the state courts.  While PCC had raised an IAC claim based on a failure to present mitigating evidence during the penalty phase, PCC had not presented this particular mitigating evidence to the state courts.*  This failure was the basis of the Martinez argument.

* Specifically, in state court, the claim focused on counsel's failure to investigate and present additional evidence regarding his tragic history of child abuse.  The new mitigating evidence in federal court was evidence of his mental illness as an adult.

This made these conservative judges angry.  In their mind, petitioner was using an improper way around Pinholster to have the federal court consider new evidence not presented to the state courts.  To them, this was not a "new" IAC claim.  In their eyes, petitioner was simply making the old claim prettier with fancy new evidence and then improperly arguing that Martinez allowed the federal court to consider that better-looking new evidence under a new IAC claim.

I find the dissenting judges' analysis a little hard to accept.  In my mind, these judges are simply unhappy with the logical consequences of Martinez.  I really don't think that there can be much of a debate that what the three-judge panel allowed here is fully consistent with Martinez -- petitioner was arguing cause based on post-conviction counsel's "errors" for failing to "properly present" an IAC claim.  Those are words quoted out of the first paragraph of Martinez.  

So Martinez has the upper hand in this round.  But the scorecard could change considerably after the en banc argument in Dickens.  Actually, reviewing the Schad unpublished order, it also appears that a Martinez argument is going to be addressed in the Detrich en banc as well.  Looks like there are going to be many battles up ahead over the reach of Martinez.

Sotomayor Writes Amicus Brief to the District Court: Hodge v. Kentucky (front-end cert denial)

By Alexandr Satanovsky

I've been meaning to post about this for a while.  On December 3, 2012, the Supreme Court denied certiorari in Hodge v. Kentucky, a front-end capital case, seeking to review the Kentucky Supreme Court's denial of a penalty phase IAC claim. The underlying facts are all too common in capital practice: Hodge's trial counsel presented no mitigation during the penalty phase; the post-conviction team developed a significant amount of truly gruesome mitigation, which was presented in a state evidentiary hearing. The State even conceded that Hodge's trial counsel was objectively deficient under Strickland. However, the Kentucky Supreme Court denied relief, finding no prejudice.

The cert. petition arose from the Kentucky Supreme Court's questionable analysis underlying the conclusion that Hodge was not prejudiced. According to Hodge, the state court's conclusion rested on a legally required "nexus" between the mitigation and the facts of the crime. Now the State disagreed with this characterization. According to the State, the Kentucky Supreme Court simply found the evidence to be factually uncompelling, in light of the aggravated nature of the homicide. You can see for yourself at pages 5-11 of the short state court opinion.

Now, I suppose that it's possible for a state court to rely, in part, on a lack of nexus, without being contrary to established law. For example, if the state court explicitly acknowledges that there is no legal significance to a (lack of) nexus, it may then find that, as a matter of fact, the nexus-free mitigation was not compelling to show prejudice in light of the aggravated nature of the homicide.  In other words, I don't think there's anything remarkable about the notion that, all things equal, mitigation without a direct nexus to the facts of the crime may be less compelling than that which does bear on the facts of the crime. [Obviously, all things are never equal when trying to compare the strengths of gruesome childhood mitigation, versus, e.g., crime-specific mitigation]. More importantly, as a matter of clearly established law, a fact-finder cannot give anything like a legal presumption--a legal thumb on the scales--to "nexus" mitigation versus "non-nexus" mitigation.

After placing the case for three consecutive conferences, Justice Sotomayor penned a lengthy dissent from the denial of cert (longer than the entire reasoned state court opinion). To her, Hodge was right that the Kentucky Supreme Court's prejudice inquiry involved a legal nexus requirement. She would have summarily vacated and remanded this case to apply the proper legal standard, allowing the state court to re-do its prejudice determination without giving legal (in)significance to Hodge's powerful childhood evidence.

This sleeper dissent seems noteworthy to me for another reason. We often hear of circuit dissents as amicus briefs to the Supreme Court; well, here, it is Sotomayor who is essentially writing an amicus brief to the Kentucky federal district judge (who will hear this claim next), stating that 2254(d)(1)'s "contrary to" clause is satisfied by the state court's improper prejudice standard. She really minces no words: "[t]he Kentucky Supreme Court’s opinion is plainly contrary to these precedents [rejecting the "nexus" requirement for mitgiation evidence]," (slip op. at 8). In another paragraph, she notes that the state court's view "that in cases involving 'violent and cruel murders, it does not matter that the 'malefacto[r]' had a 'terrible childhoo[d]' . . . is contrary to our cases applying Strickland’s prejudice prong." (My emphasis).

This strikes me as an unusual occurrence. The universe of claims that have satisfied the onerous post-AEDPA 2254(d)(1) standard is relatively small, and from this it follows that those claims should be the strongest at the front-end cert. stage, and so most likely to have been the chosen issues in the front-end cert petition. Still, none of the high profile 2254(d)(1) grants that I have read, to my knowledge, involved written dissents from the denial of state court cert., assessing the same legal question without the additional (d)(1) deference. Justice Sotomayor's opinion, in other words, is my first encounter with a dissent from a denial of cert. on a front-end claim that necessarily will be relitigated in federal court, but now with the the added cloak of 2254(d)(1). And her dissent may prove quite imporant if Hodge gets through this cloak, and wins penalty phase relief in the Sixth Circuit.

Habeas Corpus - Detrich En Banc Argument

By JK

I wish I had caught this before it happened, but the Ninth Circuit reheard the capital habeas case of Detrich v. Ryan en banc on Tuesday (12/11).  I discussed Detrich a little bit in this post.  The issues in the case, according to the rehearing petition, are:

1. Did the panel err by substituting its judgment for that of the state post-conviction court and concluding that the state court unreasonably applied Strickland v. Washington, 466 U.S. 668 (1984)?

2. Did the panel err by considering evidence that was not presented in state court notwithstanding the United States Supreme Court’s directive in Cullen v. Pinholster, 131 S. Ct. 1388 (2011), that federal review of a claim addressed on the merits in state court is limited to the evidence presented in state court?

Obviously, I care more about the second question than the first.  I am guessing the en banc court will focus more on it as well.  

To note, here's the panel: Kozinski, Pregerson, Reinhardt, Graber, W Fletcher, Gould, Bea, Murguia, Christen, Nguyen, Watford.  As I have discussed before, there is no guarantee that any of the original panel members sit on the en banc panel.  And, indeed, only one of the original members, Pregerson, is sitting on the en banc panel.  He was in the majority, which granted the habeas petition. 

Here is the 55-minute video of the en banc argument.  Haven't watched it yet, but I will. And here's the audio if you just want to listen to it.  Gotta love the Ninth Circuit for making these available (yes, I am looking right at you, Second Circuit).

 

Habeas Corpus - JK's First Ninth Circuit Post

By JK

I was hoping that my first Ninth Circuit post was going to be more substantive.  Actually, I have been working on a broad-based substantive post, but it's not finished yet.

In the meantime, I wanted to briefly discuss an en banc decision from the Ninth Circuit the other day denying habeas relief in a death penalty habeas case, Carrera v. Ayers.  It's an odd decision that will pretty much have no impact whatsoever on the future of habeas law.  How can that be the case?  Well, for one thing, it is a pre-AEDPA case.  Not many of those left out there, I reckon.  But it's not just a pre-AEDPA case, it's a pre-Batson case.  And the issue in the case is whether counsel was ineffective for failing to challenge the prosecution's use of peremptory challenges in a racially discriminatory manner.  Due to its age, the Ninth Circuit had to analyze the case under the pre-Batson standards that existed in California.  So the case is only relevant to pre-AEDPA petitions raising pre-Batson racial discrimination claims out of California.  I would be surprised to learn that any of those still existed in habeasland.

If you are interested in learning why a case in such a procedural posture still exists, the district court's decision provides the procedural history (it's available at 2008 WL 681842).  I was hoping that it was going to be somewhat interesting, but it really isn't.  Just a lot of back and forth between federal and state court and what is best described as a piecemeal resolution of the habeas petition.  On the positive side, it does appear that the district court granted partial habeas relief back in 2004 and vacated the death sentence.  The only issues on appeal were guilt-phase claims.

Not much more to say about this other than I have found that there is something neat in the way that the Ninth Circuit publishes its opinions.  A published opinion often has a "Summary" at the beginning that, according to a footnote in the summary, "constitutes no part of the opinion of the court. It has been prepared by court staff for the convenience of the reader."  That's new to me.  Obviously, when you go look at published opinions, they have the introductory paragraph that's not a part of the opinion.  But this is something a little different; it's more substantive. It's more like the syllabus at the beginning of a Supreme Court opinion.  Not as long and detailed as those, but it's definitely more than the typical introductory paragraphs I have seen out of the Second Circuit.  I'll post the summary below the fold so you can see what I am talking about.  I'll also include an introductory paragraph from a recent Second Circuit opinion to show the difference.

 

Habeas Corpus - The Rumors of Habeas's Demise Have Been Greatly Exaggerated (By Me)

By JK

Following up on Alan’s first post from the other day, I do want to make a little bit of noise around here about several good results in habeas cases in the circuit courts over the past six weeks (since around mid-September).

There have been habeas grants out of the:

Second Circuit: Young v. Conway (see Alan's post for details)

Third Circuit: Garrus v. Secretary of Dep't of Corr. - Apprendi violation

Seventh Circuit: Harris v. Thompson - right to present a defense and IAC

Ninth Circuit: Cudjo v. Ayers - violation of right to present a defense; Stankewitz v. Wong - IAC during sentencing phase of death penalty case

There may have been others; these are just those that I have seen.  And I am sure there will be more.  But I have expressed a lot of doom and gloom around here over the past two years (ever since Richter).  So I do think it’s worthwhile to acknowledge when the good things happen.  And if you have time please read the Seventh Circuit’s decision.  It’s looong, but it's pretty great.  It kind of feels like the court was convinced of petitioner's innocence, and this belief ends up being the motivating force behind the analysis. 

And the Third Circuit’s decision is pretty cool, too.  It’s an en banc decision on an Apprendi issue – the scope of the “prior conviction” exception to the Apprendi rule.  Two things of note (at least related to the blog): the majority has a great footnote (#8) that criticizes the dissent for reading the unreasonable application/clearly established law part of the 2254(d) standard in such a narrow way that it basically turns it into the “contrary to” clause.  I have been yelling about court’s doing that since nearly the inception of this blog (that post was written in July 2009; I started the blog in April 2009).  So glad to see a court say it.

At the other end, the dissent points out that the majority did not give enough deference under the standard of review and gives lists of all of the times that the Supreme Court (i) granted cert. on the issue; and (ii) reversed on this ground (both summary reversals and after merits briefing).  These lists are long.  The lack of deference summary reversal was the subject of my most recent article (part of my recent doomistry* and gloomery*).  And it does serve as a reminder that, even with the good stuff, there are still reasons to be concerned about the future of habeas.  

*Not actual words.