The court of appeals recently addressed an issue that has divided courts elsewhere: whether defense counsel may present an insanity defense without the defendant’s consent. The court ruled that defense counsel may not do so, stating that “because the decision of whether to plead not guilty by reason of insanity is part of the decision of what plea to enter, the right to make that decision is a substantial right belonging to the defendant.” Continue reading
Tag Archives: effective assistance of counsel
In State v. Harbison, 315 N.C. 175 (1985), the North Carolina Supreme Court held that when defense counsel admits the defendant’s guilt to the jury without the defendant’s consent per se ineffective assistance of counsel occurs. The Harbison Court reasoned that when counsel admits guilt without consent, it is essentially the same as entering a guilty plea on the defendant’s behalf without the defendant’s consent. It concluded: “ineffective assistance of counsel, per se in violation of the Sixth Amendment, has been established in every criminal case in which the defendant’s counsel admits the defendant’s guilt to the jury without the defendant’s consent.” Id. at 180.
On Halloween, I was dressed up as a sheep, trick or treating with my daughter, Little Bo Peep. Fortunately, serious legal business was being conducted elsewhere: the Supreme Court heard oral argument in Lafler v. Cooper, a fascinating ineffective assistance of counsel case. Here’s a summary of the case, courtesy of SCOTUSblog:
Cooper shot a woman, striking her several times below the waist. He was arrested and charged with assault with intent to murder, possession of a firearm by a felon, and other charges. The state offered a plea bargain under which Cooper could plead guilty to the assault with intent to murder charge and face a minimum sentence of fifty-one to eighty-five months in prison. Although Cooper was inclined to take the plea, his counsel advised him that, because the victim was injured below the waist, the state could not establish an element of its case, i.e., intent to murder. Based on this erroneous advice, Cooper rejected the deal. Cooper was later convicted at trial and received a sentence of between 185 and 360 months in prison.
The defendant eventually sought, and obtained, federal habeas relief based on ineffective assistance of counsel. Applying the well-settled two-pronged Strickland test for ineffective assistance of counsel, he argued (1) that his attorney performed deficiently by giving him legally inaccurate advice, which (2) resulted in prejudice when he rejected the plea offer, and ultimately received a sentence three times as long as he would have received under the deal. At least by the time the case reached the Supreme Court, the state conceded that defense counsel’s advice was wrong and that the lawyer performed deficiently, but argued that the defendant had suffered no prejudice, because he had received a fair trial and had relinquished no rights based on the faulty advice.
Over at Crime and Consequences, a prosecution-oriented blog, the thinking is this: a fair trial is the benchmark way of adjudicating criminal cases. A defendant who receives such a trial can’t complain about the result. Perhaps he didn’t get a windfall — a reduced sentence obtained through plea bargaining — but there is no right to such a windfall. (For more detail, look at the posts here and here.) This point of view can be summed up in a question asked by Justice Kennedy during oral argument: “You are saying it was unfair to have a fair trial?”
The defense counters that most cases are resolved by plea bargaining, making plea bargaining a critical stage of any criminal case, a point recently acknowledged by the court in Padilla v. Kentucky. On this view, prejudice is measured not just by whether the defendant relinquished rights based on counsel’s deficient performance, but by whether the defendant received a longer sentence than he otherwise would have received. The defense contends that “the approach advocated by the state [in Lafler] has been rejected by twelve federal courts of appeals and twenty-five of the twenty-seven states in which the issue has arisen.” (I couldn’t quickly find a case directly on point in North Carolina, but in State v. Simmons, 65 N.C.App. 294 (1983), the court of appeals awarded a new trial to a defendant whose lawyer failed to communicate a plea offer to him, perhaps implicitly rejecting the idea that a defendant who receives a fair trial can never show that he was harmed by poor lawyering during plea negotiations. If you know of a case more closely on point, please let me know.)
Lafler is an interesting case. Even if the defendant prevails, there’s a question about the proper remedy. Is the defendant entitled to specific performance of the plea bargain? Is he entitled to choose whether to accept the plea offer, and to turn it down in favor of a new trial if, for example, key witnesses have died or moved? Or should his conviction simply be set aside, with the parties free to negotiate anew or to proceed to trial as they please?
The New York Times covers the case here. The oral argument transcript is available here. I should add that the case was heard along with a companion case, Missouri v. Frye, raising similar issues on what appear to be less compelling facts.
Category: Procedure, Uncategorized | Tags: cooper, effective assistance of counsel, ineffective assistance of counsel, lafler, plea bargaining
I was catching up on the Fourth Circuit’s recent opinions this weekend when I came across United States v. Luck. It raises some interesting issues that are not specific to federal court, so I thought I’d put together a post about it.
The defendant in Luck was charged with drug and gun crimes. The government’s evidence came primarily from two informants: one was working off charges, while the other was working to help the first, and for money. Although one of the informants made two supervised drug purchases from the defendant, she was not searched thoroughly prior to the purchases. There was virtually no other corroboration of the informants’ testimony. Nonetheless, the defendant was convicted and sentenced to 444 months in prison.
The case was affirmed on direct appeal. The defendant then sought collateral review, arguing in pertinent part that his trial lawyer was ineffective for failing to request a jury instruction on informant testimony, i.e., an instruction that informants’ testimony “must be examined and weighed by the jury with greater care than the testimony of an ordinary witness” because of their particular interests and biases. The federal district court denied relief on this claim, but the Fourth Circuit reversed, vacating the defendant’s convictions.
The court began by suggesting, but not actually holding, that such an instruction should be given as a matter of course in many cases: “Among the other circuits that have considered this question, there is a consensus that an informant instruction is necessary when the informant’s testimony is uncorroborated by other evidence.” Although the court found this consenus “persuasive,” implying that judges should give such instructions sua sponte in some instances, it held only that defense counsel performed deficiently in not requesting such an instruction in the case at bar, given that counsel’s strategy focused on the informants’ motives and lack of credibility.
The court then discussed whether the defendant was prejudiced by counsel’s failure to request the instruction. It concluded that the district court likely would have given the instruction if requested, and that even though the district court gave a general instruction on witness credibility, including bias, an informant instruction would have been “more effective[]” and would have “alert[ed] jurors to the potentially unique problems that inhere where an individual is paid to inculpate a defendant.” Because the government’s case depended so heavily on the informants’ testimony, the court found a reasonable probability that an informant instruction would have changed the outcome of the trial. One judge dissented on this point; he would have held that even if counsel’s performance was deficient, there was no prejudice because the jury was clearly aware, through defense counsel’s cross-examination and closing argument, of the informants’ motives.
Turning to North Carolina law, there is, of course, a pattern jury instruction for informants. N.C.P.I. Crim — 104.30, entitled “Informer or Undercover Agent,” reads as follows:
You may find from the evidence that a State’s witness is interested in the outcome of this case because of the witness’ activities as an [informer] [undercover agent]. If so, you should examine such testimony with care and caution in light of that interest. If, after doing so, you believe the testimony in whole or in part, you should treat what you believe the same as any other believable evidence.
A couple of asides. First, it strikes me as odd that this instruction covers undercover officers as well as informants, given that officers generally are not considered interested witnesses. State v. McQueen, 181 N.C. App. 417 (2007). What makes undercover officers more interested than uniformed ones? However, by its terms, the instruction clearly does apply to undercover officers, and failure to give the instruction in a case involving an undercover officer was held to be reversible error in State v. Black, 34 N.C. App. 606 (1977). Second, this instruction is not as strong as the instruction discussed in Luck. It tells the jury only to view an informant’s testimony with “care,” not with “greater care than the testimony of an ordinary witness[].” That makes me wonder whether defense lawyers in some cases might reasonably request an instruction that is more robust than the pattern.
In any event, Black and other cases make it perfectly clear that the instruction must be given when requested and appropriate. Is failure to request the instruction when appropriate then ineffective assistance of counsel? I don’t think there’s a North Carolina case on point, but the reasoning of Luck is persuasive: in at least in some circumstances, the answer may well be yes.
