Ineffective Assistance of Counsel

The Sixth Amendment provides that a criminal defendant shall ‘‘have the Assistance of Counsel’’ for his defense. The Supreme Court has interpreted this clause as guaranteeing the effective assistance of counsel, beginning with one of the appeals from a trial in the Scottsboro cases (Powell v. Alabama, 287 U.S. 45, 1932). The modern standard for evaluating claims of ineffective assistance of counsel was set in 1984 in Strickland v. Washington, 466 U.S. 668.

Strickland involved an appeal from a habeas corpus petition filed by a capital defendant who claimed that his lawyer had been ineffective at trial. The lawyer did not investigate potential mitigating evidence that might have persuaded the trial judge to impose a sentence of life imprisonment, rather than death. The Supreme Court set out a two-part test for evaluating claims of ineffective assistance of counsel.

First, the defendant must show that his or her lawyer failed to provide reasonably effective assistance. The reasonableness of the lawyer’s performance is to be evaluated with reference to prevailing professional norms, and courts should be reluctant to second-guess the lawyer’s tactical judgments. Perhaps in hindsight, it appears that a lawyer should have pursued a particular line of investigation or tried a different strategy at trial, but if the decision was reasonable under the circumstances at the time it was made, the lawyer’s performance should not be judged ineffective. The Court declined to adopt a set of rules for the evaluation of counsel’s performance at trial out of concern for restricting the flexibility a defense lawyer must have in making tactical decisions. It also emphasized that review of counsel’s performance should be highly deferential, with a heavy presumption in favor of a finding of effective representation.

The second prong of the Strickland test is that the defendant will not be entitled to reversal of the conviction unless the lawyer’s deficient performance resulted in prejudice to the defendant. To make a showing of prejudice, the defendant must establish that, except for the lawyer’s errors, there is a reasonable probability that the result of the proceeding would have been different. As the Court recognized, it is possible to avoid the first prong of the test entirely and resolve a case on prejudice. If the evidence of guilt against the accused is overwhelming, it is unlikely that even a serious error of counsel would change the result of the proceedings. Following Strickland, lower federal courts frequently resolved ineffective assistance claims on the prejudice prong.

In certain cases, prejudice is presumed. The most significant category of these cases involves the attorney’s representation of conflicting interests. If a conflict of interest actually affects the lawyer’s performance, the defendant need not show prejudice to obtain a new trial (Cuyler v. Sullivan, 446 U.S. 335, 1980). The requirement of showing actual effect means that the defendant must specifically identify some act or omission that is traceable to the lawyer’s multiple loyalties. Courts have been unwilling to expand the categories of cases in which prejudice is presumed, and the Supreme Court has actually suggested that only the concurrent representation of clients with conflicting interests should trigger the presumption of prejudice (Mickens v. Taylor, 535 U.S. 162, 2002).

The Court’s approach to ineffective assistance of counsel has been criticized on a number of grounds. First, by deferring to the norms prevailing in the practicing bar, the Court has abandoned any attempt to improve the standards of practice of criminal defense lawyers. If the overall performance of the criminal defense bar is deficient, a given lawyer’s performance may not violate the first prong of Strickland.

Because of the overwhelming caseloads of many public defender offices and lack of adequate funding for indigent defense, it is possible that the average level of competence of lawyers who represent poor people may not be very high. Several notorious cases involving defense lawyers who slept through substantial portions of trials in capital cases brought this problem to public attention. Two recent Supreme Court decisions seem responsive to this concern and may reinvigorate the efforts of courts to set minimum standards for the competence of criminal defense lawyers (Wiggins v. Smith, 539 U.S. 510 (2003); Rompilla v. Beard, 545 U.S., 2005). Both cases involved inadequate investigations of mitigating circumstances performed by lawyers representing capital defendants.

A related criticism was raised by Justice Marshall, dissenting in Strickland. Marshall asked why the rights of the accused depend upon the outcome of the proceeding. Should not the defendant’s rights be defined instead in terms of ‘‘vigorous and conscientious advocacy by an able lawyer,’’ regardless of whether the defendant is likely to be convicted anyway? Putting it another way, the defense lawyer’s role is subordinated to the systemic goal of convicting guilty people, rather than protecting the intrinsic procedural rights of the accused.

Another concern with Strickland relates to the prejudice prong. Any single error by counsel is unlikely to have a direct causal relationship with the outcome of the trial. Instead, it is more likely that a series of small errors adds up to a different result. It can also be difficult for a defendant to prove a negative. The trial record will reveal the result of counsel’s actual performance, but will offer no clue as to the result of a hypothetical decision. If the lawyer had filed a suppression motion, investigated a line of mitigating evidence, or called a different witness, the result would undoubtedly be different. From the standpoint of a reviewing court, however, it is exceedingly difficult to say how, exactly, the result would have been different. Moreover, the record will be affected by the defense lawyer’s incompetence, making it even more likely that prejudice will not be apparent.

Barring a major change in the funding for indigent representation, ineffective assistance of counsel claims are likely to persist. Courts must balance the desirability of enhancing the performance of lawyers against the recognition of the difficulty of establishing ineffectiveness upon appellate and collateral review. The difficulty of this balance is exacerbated by the complexity of the defense lawyer’s task and the difficulty of judging tactical decisions in hindsight. Strickland may not be entirely satisfactory, but it remains the constitutional framework within which these decisions will be made.

W. BRADLEY WENDEL

References and Further Reading

• Bright, Steven B., Counsel for the Poor: The Death Sentence not for the Worst Crime but for the Worst Lawyer, Yale Law Journal 103 (1994): 1835–1883.
• Burkoff, John M., and Hope L. Hudson. Ineffective Assistance of Counsel. St. Paul, MN: West Group, 2002 and Supp. 2004.

Cases and Statutes Cited

• Cuyler v. Sullivan, 446 U.S. 335 (1980)
• Mickens v. Taylor, 535 U.S. 162 (2002)
• Powell v. Alabama, 287 U.S. 45 (1932)
• Rompilla v. Beard, 545 U.S. (2005)
• Strickland v. Washington, 466 U.S. 668 (1984)
• Wiggins v. Smith, 539 U.S. 510 (2003)

See also Habeas Corpus: Modern History; Right to Counsel; Rights of the Accused; Scottsboro Trials