A Good Week for the Right to Counsel

With three decisions this week, the U.S. Supreme Court has signficantly strengthened defendants’ right to effective assistance of counsel.  In my next post, I’ll cover the two decisions that have received the most press coverage, Lafler v. Cooper and Missouri v. Frye.  Here, I’ll comment on Martinez v. Ryan, a decision that seems much narrower than the other two, but that may ultimately prove no less significant.  At issue is the right to counsel in collateral proceedings — an issue that has been hotly contested at the Court for many years.

Here’s what happened.  An Arizona jury convicted Martinez of sexual conduct with a minor.  After trial, the state appointed a new lawyer to handle Martinez’s post-conviction challenges.  This lawyer pursued a direct appeal, which proved unsuccessful, and also initiated a state collateral proceeding by filing a “Notice of Post-Conviction Relief.”  Later, however, the lawyer filed another statement asserting that she could find no colorable claims for collateral relief.  Martinez then had 45 days to file a pro se petition in order to keep the collateral proceeding alive.  He did not do so, allegedly because his lawyer failed to advise him of what was happening in his case.  In any event, the state trial court then dismissed the collateral proceeding.


A year and a half later, Martinez — now represented by a different lawyer — filed a second notice of post-conviction relief in state trial court.  This time, he also presented a claim for relief, asserting that his trial lawyer had violated his constitutional right to effective assistance of counsel by, for instance, failing to call an expert witness to rebut certain expert testimony used by the state.  The claim was dismissed, however, on procedural grounds because it could have been raised in the first collateral proceeding.  A subsequent federal habeas petition was likewise dismissed because of the state procedural default.  Although federal habeas law recognizes an exception to the procedural default bar when a petitioner shows “cause and prejudice,” the lower federal courts held that counsel errors during collateral proceedings may never satisfy the “cause” part of the test.  Indeed, the Supreme Court had indicated as much in Coleman v. Thompson, 501 U.S. 722 (1991).

In Martinez, though, the Court recognized an exception to Coleman and held that in certain circumstances errors in collateral proceedings might constitute adequate cause to excuse a procedural default.  More specifically, key to the Court’s holding in favor of Martinez was the fact that, under state law, he was only permitted to challenge the competence of his trial counsel in a collateral proceeding.  Thus, poor representation in the collateral proceeding would effectively torpedo his one and only opportunity to present the ineffective assistance claim in state court.

Indeed, there is much that could be said in favor of recognizing a constitutional right to counsel (including effective assistance of counsel) in collateral proceedings in such circumstances.  The Constitution guarantees a right to counsel on direct appeal from a conviction; if there is an issue that cannot be raised on direct appeal and must await a collateral proceeding, then the same reasoning that supports a right to counsel on direct appeal would also seemingly support a right to counsel in that collateral proceeding.  However, the Court has steadfastly declined to recognize a right to counsel in collateral proceedings, and the Martinez Court did not take advantage of the opportunity it had to carve out an exception to the no-counsel rule.  Yet, in opening up federal habeas review under the cause-and-prejudice test in cases like Martinez, the Court has gone a long way toward achieving the same practical consequences that would come from recognizing a right to counsel in these circumstances.

So, how is Martinez good news for defense representation when the Court ended up not formally adopting a right to counsel?  First, the Court opened up new opportunities in federal habeas for some defendants to challenge the quality of their trial counsel.  Second, by making the quality of collateral counsel relevant in some cases, the Court has created new incentives for states to ensure that collateral counsel is competent and to provide remedies when collateral counsel is not up to snuff.

To be sure, Martinez could be read narrowly to apply only in states like Arizona that preclude ineffective assistance claims on direct appeal.  But there is a potential for Martinez to reach more broadly.  For instance, what about cases in which the same lawyer represents a defendant at trial and on direct appeal?  Appellate counsel cannot ethically challenge his own competence, so any claims of ineffective assistance must as a practical matter await collateral proceedings.  There seems no good reason not to extend Martinez to that setting.  Likewise, where the factual basis of an ineffective assistance claim is not known until after the direct appeal, Martinez should also apply.

More generally, there may be an argument that Martinez should apply whenever the legal or factual basis of any type of claim is not reasonably available during the direct review process.  For instance, Brady violations often don’t come to light until after direct appeals.  If an incompetent collateral counsel causes a late-discovered Brady claim to be defaulted in state court, then that incomptence should count as “cause” to excuse the default and permit a federal habeas court to consider the Brady violation on the merits.

An interesting question is how this reasoning should play out for defendants who represent themselves in collateral proceedings.  Assume, for instance, that Martinez requested, but was denied, a court-appointed lawyer for his first round of collateral review.  If it was Martinez then, and not a lawyer, who neglected to make a claim of ineffective assistance in the first collateral proceeding, might the “cause” standard for federal habeas review nonetheless be satisfied?  Is there any reason to treat differently the defendant who is stuck with an incompetent lawyer and the defendant who, recognizing his own limitations, tries and fails to get a lawyer?  Either way, it seems to me that the errors in representation are, as a practical matter, equally beyond the defendant’s control.  Also, if Martinez does not extend to pro se defendants, states would have perverse incentives to restrict the availability of court-appointed counsel in collateral proceedings.  It is possible that Martinez might even end up being a net negative development for defendants.

In any event, it was interesting for me to read Martinez the same week that I taught Harrington v. Richter, 131 S.Ct. 770 (2011), in my Post-Conviction Remedies class.  Harrington, which is barely a year old — see post here — sent a strong message against federal habeas courts granting ineffective assistance claims.  Remarkably, the author of Harrington was also the author of Martinez, Justice Kennedy.  What gives?  In 2011, Justice Kennedy is chewing out the Ninth Circuit for granting an ineffective assistance claim, but in 2012 Kennedy is overturning the Ninth Circuit’s denial of such a claim.  Aside from a general antipathy to the Ninth Circuit, what could reconcile these decisions?

The key difference may be that Harrington dealt with the application of 28 U.S.C. 2254(d), which mandates a high level of deference to state-court decisions on the merits, while Martinez dealt with a claim that the state courts expressly declined to address on the merits.  The lesson here may be that Justice Kennedy, a swing voter in habeas cases, is more inclined to defer to state-court decisions on the merits than state-court decisions that don’t reach the substance of a defendant’s claims.  Such a distinction would, I think, be quite sensible.  Habeas law should encourage state courts to take defendants’ claims seriously.  The procedural default rule of Wainwright v. Sykes — a judicial invention not mandated by statute — does just the opposite.  I’m happy to see the Martinez Court expand the cause-and-prejudice exception to the procedural default rule.  Better still would be a complete abandonment of Wainwright, which is no longer really necessary to advance finality and comity interests in criminal litigation in light of the later-adopted total-exhaustion rule of Rose v. Lundy and the many restrictions on federal habeas imposed by the Antiterrorism and Effective Death Penalty Act of 1996, including the super-deference rule of § 2254(d).