Deficient Performance

The first prong of the Strickland test asks whether counsel made errors so serious that counsel was not functioning as the "counsel" the Sixth Amendment guarantees.¹ The question concerns the quality of the lawyer's work, separate from what the work produced. The yardstick is reasonableness under prevailing professional norms, applied objectively.² A defendant who satisfies this prong has shown only that the representation was unreasonable. Whether the unreasonable conduct cost the defense anything is a separate question, which the prejudice prong answers later in this chapter.

The shape of the standard follows from the position of the court that applies it. Every defendant who raises this claim has already lost, and the reviewing court reads the record knowing how the case ended. A test that measured lawyers by results, or by what hindsight says would have worked, would condemn nearly every unsuccessful defense, because a loss makes each road not taken look promising. Strickland therefore directs courts to eliminate the distorting effects of hindsight and to judge counsel's conduct from counsel's perspective at the time.³ Reasonable lawyers also defend the same case in different ways, so the standard leaves wide room for professional judgment and condemns only conduct that falls outside what a competent lawyer might have done.

Three ideas make up this prong, and each receives its own subsection. The first is the objective standard itself, which supplies the measuring stick. The second is the strong presumption of competence and the ban on hindsight that surround that standard and set the lens through which a court looks at counsel's choices. The third is the duty to investigate, which shows the standard at work in a concrete setting, and Strickland itself said that its general standards needed no special amplification to define that duty, the one at issue in the case.⁴ The first subsection takes up the objective standard.

Objective Reasonableness

The measuring stick for the first prong is an objective standard of reasonableness, and the defendant must show that counsel's representation fell below it.⁵ Strickland describes the proper measure of attorney performance as reasonableness under prevailing professional norms.⁶ Put as a question, a court asks whether, considering all the circumstances, the acts or omissions the defendant identifies fell outside the range of professionally competent assistance.⁷ The word objective matters: the court measures the lawyer against the practices of reasonable lawyers in the relevant professional community, whatever the lawyer personally believed about the approach.⁸

The standard is anchored to professional practice because representation is an art. An act that is unprofessional in one case may be sound or even brilliant in another, so no list of forbidden conduct could tell defense lawyers in advance what to avoid. Strickland accordingly refused to adopt detailed rules for counsel's conduct. No set of rules could account for the variety of circumstances defense lawyers face or the range of legitimate decisions about how to defend a client, and rules would intrude on counsel's independence and distract from vigorous advocacy.⁹ Tying the inquiry to what the profession does supplies a benchmark from outside the courtroom. Without it, each judge would measure the lawyer against a private picture of good lawyering, and identical representation could pass before one judge and fail before another. The anchor also gives lawyers a workable target, since they can learn what the profession expects before trial instead of discovering it afterward. The formulations that preceded Strickland, such as the farce-and-mockery test and the call for normal skill and knowledge, left open what level of practice counts as customary, diligent, or reasonable, and reference to prevailing professional norms answers that open question by pointing to the profession itself.¹⁰

Prevailing norms come from the profession's own practices, and a court may consult bar association standards, defender-organization performance guidelines, and practice manuals to learn them. In Padilla v. Kentucky the Court treated such sources as guides to what is reasonable, and as guides only, not as commands that settle the question.¹¹ The rules of professional conduct occupy the same place: they can be relevant evidence of prevailing norms, but they do not themselves define the constitutional floor.¹² The reason lies in what the Sixth Amendment guarantees. The guarantee exists to ensure a fair trial, and improving the quality of legal representation is a separate goal, however valuable.¹³ A disciplinary rule or an aspirational guideline may demand more than a fair trial requires, so a lawyer's departure from one does not show that the lawyer stopped functioning as counsel. Before Strickland, some lower courts had treated a substantial departure from generally accepted guidelines as automatic proof of incompetence, and the Court rejected that approach.¹⁴ Students often treat a departure from a written guideline as proof of deficiency, when it is only evidence of what the norm is.

The guideline was identical in both cases, so it cannot explain the different results. What moved the answer was whether the omission fit the judgment a reasonable lawyer would make in those circumstances, which is the question the standard asks.

The standard also does not demand a perfect lawyer. A defendant is entitled to reasonably competent advice, not flawless advice, because even excellent counsel cannot predict with certainty how a court will rule or which witnesses a jury will believe.¹⁵ Reasonable lawyers defend the same case in different ways, so work that is merely below average can still sit inside the range of professionally competent assistance, and the prong asks only whether the conduct fell outside that range.¹⁶ Courts find deficiency most readily when the record shows abdication rather than an exercise of professional judgment, meaning that the lawyer overlooked something instead of choosing among options.¹⁷ In Kimmelman v. Morrison the lawyer failed to file a motion to suppress within the statutory deadline and excused the failure by saying he did not know the evidence had been seized. The Court found ignorance of the law so basic that no strategic judgment could explain it, and held that the conduct fell outside prevailing professional norms.¹⁸

Two decisions show the standard operating at opposite ends of its range. In Padilla, a long-time lawful resident pleaded guilty to a drug-distribution charge after counsel's advice failed to account for the deportation consequence, and the Kentucky Supreme Court held that deportation was a collateral consequence that gave no basis for relief.¹⁹ The Court found the direct-versus-collateral distinction ill-suited to deportation, which is so closely tied to the criminal process that it is hard to separate from the conviction.²⁰ It then asked what the profession expected and found that the weight of prevailing professional norms, reflected in defender guidelines, bar standards, and practice manuals, required counsel to advise a client about the risk of deportation.²¹ Because the immigration statute stated in plain and explicit terms that the conviction would lead to deportation, counsel's failure to say so fell below the constitutional standard.²² Padilla decided only the performance question and left prejudice open.²³ Where the immigration consequences are less clear, counsel must at least advise that the plea carries a risk of deportation, and courts remain divided over whether a comparable duty reaches other collateral consequences such as mandatory sex-offender registration.²⁴

Nix v. Whiteside sits at the other end. A murder defendant claiming self-defense first said he did not see a gun, then changed his account to say he saw something metallic in the victim's hand. His lawyer warned him that he could not testify falsely and that, if he insisted on the new version, the lawyer would tell the court of his belief that the testimony was false. Assuming the lawyer knew the account was a lie, as the Court assumed, that response was within the range of acceptable professional responses and therefore was not deficient performance under the first prong.²⁵ The decision does not say the lawyer had to respond that way. Some states follow a different approach, a free narrative in which the lawyer invites the client to tell what happened without directed questions, and a lawyer who used it would also be acting within the range.²⁶

What separates the two cases is the shape of the norm each one applied. In Padilla the profession's practice converged on a required act, so omitting it landed outside the range. In Nix the profession tolerated more than one response, so the lawyer's choice of one landed inside it. The standard therefore marks the outer edges and leaves the interior open: a lawyer can be faulted for omitting what competent lawyers uniformly do, and cannot be faulted for choosing one of several acceptable paths. The next subsection takes up the lens a court uses in drawing that line, a strong presumption that counsel's conduct falls inside the range together with a ban on hindsight.

Presumption of Competence

The objective standard marks where the line between competent and deficient representation falls, and a court still needs a way to find that line on an actual record. Strickland supplies the lens: judicial scrutiny of counsel's performance must be highly deferential, and a court must make every effort to eliminate the distorting effects of hindsight, to reconstruct the circumstances of the challenged conduct, and to evaluate it from counsel's perspective at the time.²⁸ Within that lens the court begins from a presumption of competence, a strong presumption that counsel's conduct falls within the wide range of reasonable professional assistance, and the defendant must overcome the presumption that the challenged action might be considered sound trial strategy.²⁹ The same idea appears in the controlling statement that effective assistance is presumed unless counsel's conduct so undermines the adversarial process that the trial cannot be relied upon to have produced a just result.³⁰ The word might in the strategy formulation sets the question: whether the action is the kind of choice that sound strategy could produce, which is a lower hurdle for the lawyer than proof that the choice was the best one available.

The presumption rests on three separate reasons, and each guards against a different risk. The first is hindsight. Every defendant who raises this claim has lost at trial, and a reviewing court reads the record knowing how the case ended, so every road not taken looks more promising than it did when counsel had to choose.³¹ Deference corrects that skew by making the court ask what counsel could see at the moment of decision. Students go wrong when they treat the failure of a tactic as evidence that the tactic was unreasonable, because reasonable tactics fail all the time.

The second reason is the variety of legitimate choices. There are countless ways to provide effective assistance in a given case, and even the best defense lawyers would not defend the same client in the same way.³² A court that treated every difference in judgment as error would be substituting its own preferred strategy for counsel's, and lawyers who expect that kind of review have a reason to pick the defense easiest to justify later over the one best for the client. The same concern for counsel's independence led Strickland to reject detailed guidelines for conduct.³³ The third reason is the cost of the remedy. A finding of deficiency that caused prejudice generally means a new trial with substitute counsel, an intrusive result, so courts tend to bend over backwards to find that performance was not deficient.³⁴ A system handling this many convictions could not absorb retrials whenever a losing defense looked improvable in retrospect, which is why the claim is raised constantly and rarely wins.³⁵

Strickland's own sentencing claim shows the lens at work. The defendant faulted counsel for not requesting a presentence or psychiatric report, not investigating and presenting character witnesses, and not cross-examining the medical experts.³⁶ Read from counsel's side, those omissions formed a plan: rely on the defendant's acceptance of responsibility at the plea colloquy and on the sentencing judge's reputation for valuing it, and argue that the crimes were committed under extreme emotional disturbance. Counsel could reasonably conclude that further character and psychological evidence would add little and would invite damaging cross-examination, and a presentence report would have exposed a criminal history more extensive than the plea proceeding had shown. The Court therefore treated the alleged deficiencies as reasonable strategy rather than neglect, and the first prong failed.³⁷

The same reasoning covers the commonplace decision to leave a witness off the stand. A lawyer who does not call an alibi witness because she believed the witness would not be credible, or because another defense seemed more likely to succeed, has made the kind of tactical judgment that courts very rarely find deficient.³⁸ The lawyer's reasons supply the sound strategy the defendant must overcome, and the defendant's later claim that the alibi would have helped is the hindsight the presumption screens out. A strategic choice made after thoughtful consideration generally will not support an ineffective-assistance claim. In Torres v. State the Georgia Supreme Court rejected such a claim where counsel weighed an insanity defense, took appropriate steps to evaluate whether it was viable, discussed it with the defendant, and decided against it.³⁹

The presumption has limits, because a presumption can be rebutted and an immunity cannot.⁴⁰ It protects choices, and the protection weakens when the record shows that no choice was made.

Both lawyers left the same witness off the stand, and what moved the answer was whether counsel made an informed judgment before doing so. A court defers to a strategy that exists; where the omission reflects no judgment, it has nothing to defer to. The presumption therefore starts every claim in counsel's favor without ending it there. How much counsel must learn before a choice counts as strategic is the question the next subsection takes up under the duty to investigate.

Exam Analysis Steps

Because the presumption controls how a court reads the record, a claim about a tactical choice is best run through the same sequence of moves each time. The steps are these:

Identify the specific act or omission the defendant challenges, rather than grading the representation by how the case ended.
Reconstruct the circumstances counsel faced when the choice was made, setting aside everything that later events revealed.
Begin from the presumption that counsel's conduct falls within the wide range of reasonable professional assistance, and ask whether the action might be considered sound trial strategy.
Ask whether the defendant has shown, from the record, that counsel made no informed choice or that the action fell outside the range of competent assistance; if not, the claim fails on the performance prong.

Duty to Investigate

The presumption of competence protects a lawyer's choices, which raises the question of how much counsel must learn before a choice counts as a strategic one. The duty to investigate supplies the answer, and it needs no separate test. Strickland v. Washington said that its standards of objective reasonableness and deferential review require no special amplification to define counsel's duty to investigate, which was the very duty at issue in that case.⁴² Competent counsel therefore generally must conduct some investigation, and a strategic choice resting on inadequate investigation can be deficient performance.⁴³ The Court did not go to the opposite extreme either: a reasonable decision not to investigate some aspect of the case can satisfy the Sixth Amendment, and a failure to pursue a point that later turns out to matter is not deficient performance by itself.⁴⁴ A court therefore judges the reasonableness of the decision whether to look, as counsel could see it at the time, and does not ask what the investigation might later have revealed.

The two halves of that rule exist for different reasons. The requirement that counsel do some investigation follows from what deference is for. The presumption protects a lawyer's judgment between options, and a judgment needs raw material: a lawyer who never examines the evidence has not weighed one theory against another, so there is no choice for the court to defer to. The limit on that requirement follows from the fact that time and money are finite, which forces every defense team to decide where to look. If every lead left unexamined counted as deficient, any fact that surfaced after a loss would become a constitutional violation, and that is precisely the hindsight distortion the deferential lens is built to remove. Asking whether the decision to look or not to look was itself reasonable when made keeps both pressures in view.

The idea that a decision to go forward without looking cannot be assumed to be sound is older than Strickland. In Powell v. Alabama, lawyers took charge of the defendants' cases on the morning of trial with no opportunity to prepare, and the Court held that the defendants had not received the right to counsel in any substantial sense.⁴⁵ The Court declined to assume that counsel had weighed the matter and concluded there was no defense, because neither counsel nor the trial court could say what a prompt and thorough investigation might have disclosed.⁴⁶ Powell arose from a trial court's haste, which left counsel no chance to investigate at all, while modern claims usually involve lawyers who had the time. The reasoning still carries over, because a decision made without information is not the kind of professional judgment the presumption protects.

Strickland's own facts show the deferential result. The defendant there challenged counsel's decision to present little mitigating evidence at his capital sentencing, and the Court held that counsel's choice was a reasonable strategic choice, so the first prong failed.⁴⁷ As the last subsection showed, counsel had a reason for each omission that fit a plan, so the limit on the work was itself a judgment the record could explain. Wiggins v. Smith shows where a limit on the work stops being reasonable. Counsel there ended the penalty-phase investigation after reviewing a presentence report and social services records, although standard practice in Maryland at the time included a social history report and funds were available to commission one.⁴⁸ Those records already disclosed an alcoholic mother and troubled years in foster care, which a competent lawyer would have recognized as leads toward evidence useful in choosing among possible defenses, and nothing counsel had learned suggested that the mitigation evidence would backfire or that pursuing it would be fruitless.⁴⁹ The Court held that stopping was unreasonable, because the failure resulted from inattention rather than strategic judgment.⁵⁰

Rompilla v. Beard adds that the lead can come from the prosecution. The state gave advance notice that it would use the defendant's prior rape-assault conviction to prove a history of violence, yet defense counsel did not look at that case file until shortly before the sentencing hearing and never read the rest of it, which documented severe mental illness, child abuse, and alcoholism.⁵¹ The Court held that counsel must make reasonable efforts to obtain and review such material even when the client and family members have suggested that nothing helpful exists.⁵² Until counsel has looked, counsel is not in a position to make a reasonable strategic choice.⁵³ Set beside Strickland, the two cases differ in the information behind the limit. In Strickland the limit followed a plan counsel could explain. In Wiggins counsel had leads in hand, the means to pursue them, and no stated reason for stopping, and in Rompilla the opposing side had pointed to the file that made reviewing it the obvious first step.

The same line runs through noncapital cases, including defenses based on mental state. Failing to consult experts about the defendant's mental functioning or to secure the defendant's medical records can be ineffective assistance.⁵⁴ In People v. Graham, a New York appellate court found that a defendant convicted of murdering his former girlfriend had been denied meaningful representation, because the prosecution's case turned almost entirely on his state of mind and counsel had not taken the minimal steps of obtaining psychiatric records and arranging an expert evaluation, which were needed to decide whether to present a psychiatric defense.⁵⁵ The court framed the question as whether failing to secure and review crucial documents amounted to meaningful representation, rather than whether leaving them out was a legitimate tactic.⁵⁶ The Georgia decision in Torres v. State, mentioned in the last subsection, comes out the other way on a similar defense: counsel considered an insanity defense, took appropriate steps to evaluate its viability, and discussed it with the defendant before deciding against it, so the choice rested on information and was not ineffective assistance.⁵⁷

Both lawyers left something unexamined, and what moved the answer was whether counsel looked before deciding. The first lawyer's theory has no information behind it, so the deference owed to an informed strategy never arises and the performance can be found deficient. The second lawyer's limit has a stated basis in the file, which is the kind of reasonable decision not to investigate a particular point that the Sixth Amendment tolerates.⁵⁸

Exam Analysis Steps

Because the claim turns on the basis for counsel's decision rather than on how the case ended, it is best run as a sequence that reconstructs what counsel knew. The steps are these:

Identify exactly what counsel did not investigate and what counsel did instead, whether that was a chosen defense theory or a limit on the work.
List the leads available to counsel at the time, including records, reports, the client's own statements, notice from the prosecution about its evidence, and the standard practice in that jurisdiction.
Ask whether a reasonable lawyer holding those leads would have pursued them, taking account of limited time and resources and of any sign that the lead would be fruitless or harmful.
Decide whether counsel actually made an informed decision about the lead or simply never looked, remembering that the record of inattention supports deficiency and a reasoned limit supports deference.
Set aside the later discovery that the uninvestigated point proved important, because that fact bears on prejudice and does not by itself show deficient performance.

Students tend to err in both directions: some treat the word strategy as a password that ends the inquiry, and others treat any fact found after a loss as proof that the lawyer should have looked. Step four tests whether a strategy exists behind the label, and step five keeps hindsight out of the performance question.