New Jersey Leibman Response



Prepared by:

Bennett A. Barlyn

Deputy Attorney General

Division of Criminal Justice

Appellate Bureau

November 2000


“A Broken System: Error Rates In Capital Cases, 1973-1995″ published by Professor James S. Liebman of the Columbia Law School and several colleagues in June 2000 purports to track every death sentence case that went through the legal system in the 23 years following the United States Supreme Court’s 1972 landmark decision in Furman v. Georgia, 408 U.S. 238 (1972) which held that the existing practice of absolute jury discretion in capital sentencing resulted in the arbitrary and discriminatory infliction of the death penalty in violation of the Eight and Fourteen Amendments.

The study had its origin in a request by Senator Joseph F. Biden, Jr., then chairman of the Senate Judiciary Committee, in 1991 to Professor Liebman to calculate the frequency with which federal judges found errors in appeal of death penalty cases and then set aside the sentence.

Professor Liebman derived information used for the study from a search of all published state and federal judicial opinions in the United States conducting direct and habeas review of state capital judgments. He then 1) checked and catalogued all the cases the opinions revealed; and 2) collected hundreds of items of information about these cases from published decisions and the NAACP Legal Defense Fund’s quarterly death row census; and 3) tabulated the results.

Notably, the study did not include New York, New Jersey, and Connecticut, states in which either no cases have been appealed or no appeals have gone through the full three-stage (direct appeal, state post-conviction review, and habeas) review process.

The principal findings set forth in the study are as follows: of the 4,578 death sentences adjudicated completely, i.e., through federal habeas review, during the 23-year period, 68% — more than two out of three — were found to be “seriously flawed.” According to the study, 1,885 death sentences (41%) were reversed because of serious error when reviewed on direct appeal. Of the death sentences that survived state direct and post-conviction review, 599 were finally reviewed in a first federal habeas corpus petition during the 23-year study period. Of those 599 death sentences, 237 (40%) were reversed due to serious error. Based on the foregoing, the study concludes that “[n]ationally, over the 1973-1995 period, the overall error-rate in our capital punishment system was 68%.”

The study identified the most common errors necessitating reversal as (1) egregiously incompetent defense lawyering (accounting for 37% of the state post-conviction reversals), and (2) prosecutorial suppression of evidence that the defendant is innocent or does not deserve the death penalty. In the three states with the most executions since 1976, error rates ranged from 18% in Virginia to 52% in Texas and 73% in Florida.

According to the study, the average time between sentencing and execution was nine years. As a result, only five percent of all defendants (5,760) who had been sentenced to death since 1973 had been executed.


It is unlikely that even Professor Liebman would characterize himself as a “neutral” observer in the death-penalty debate. On the contrary, he is a zealous partisan, clearly committed to the abolition of capital punishment. Between 1982 and 1997, and most recently in Lindh v. Murphy, 521 U.S. 320 (1997), Liebman represented no less than eight capital defendants in appeals before the United States Supreme Court. Prior to joining the Columbia Law School faculty in 1985, Professor Liebman served as assistant counsel to the NAACP Legal Defense and Education Fund for six years. (Exh. 4). During his tenure with that organization he specialized in school desegregation, capital punishment, and habeas corpus matters. He is also co-author of a defense-oriented treatise on habeas practice entitled Federal Habeas Corpus Practice and Procedure.

His vehement opposition to capital punishment is plainly revealed in the opening pages of the study. It begins with a lengthy discussion of recent developments in various states, including the moratorium on executions imposed by the governor of Illinois, which Professor Liebman believes reflects a marked decline in support for capital punishment nationwide. Professor Liebman then specifically attributes this decline (unaccompanied by any supporting empirical evidence) to the fact that death sentences are perceived by the general public as “fraught with error, causing justice too often to miscarry, and subjecting innocent and other undeserving defendants — mainly, the poor and racial minorities — to execution.”[1] Doubtless, the authors of “A Broken System” are confident that their study will intensify and galvanize opposition to capital punishment by broadly indicting a system which they allege is neither a success nor even “minimally rational.”


In response to the publication of “A Broken System,” the preeminent sociologist and death penalty supporter, Professor James Q. Wilson, published an op-ed piece in the New York Times inviting readers to note what Liebman did not, nor evidently could not, claim: that, at present, there is no reliable proof that any innocent person has been executed since the resumption of capital punishment in 1973. James Q. Wilson, What Death-Penalty Errors?, New York Times, July 10, 2000. (Exh. 5). At best, according to Wilson, Liebman can only speculate that the large number of appellate reversals “leaves doubt [appellate courts] do catch” all of the errors.

Professor Wilson also astutely noted what is perhaps the report’s most conspicuous shortcoming; the fact that it “lumps together cases going back to 1973 with those decided more recently, even though the Supreme Court in 1976 created new procedural guarantees that automatically overturned many of the death-penalty [verdicts rendered] between 1973 and 1976.”

Because Professor Wilson is neither a legal historian nor a lawyer, he can be forgiven for understating the significance of the profound evolution and development of federal death penalty jurisprudence in the decade immediately following the United States Supreme Court’s 1976 decision in Gregg v. Georgia, 428 U.S. 153 (1976) and the impact these developments necessarily have had on the viability of death sentences meted out prior to and during this period. Below is small selection of seminal Supreme Court cases decided between 1976 and 1988 which unquestionably generated numerous reversals in many cases tried shortly after Furman.

Woodson v. North Carolina, 428 U.S. 280 (1976): Invalidating as cruel and unusual a death penalty statutory scheme which mandated a death sentence when the jury found defendant guilty of first-degree murder.

Coker v. Georgia, 433 U.S. 584 (1977): Holding that the sentence of death for the crime of rape was grossly disproportionate and an excessive punishment forbidden by the Eighth Amendment.

Lockett v. Ohio, 438 U.S. 586 (1978): Holding that a sentencer cannot be precluded from considering as a mitigating factor any aspect of defendant’s character or record and any circumstances of the offense defendant offers in mitigation.

Godfrey v. Georgia, 446 U.S. 420 (1980): Holding that the Georgia Supreme Court’s broad and vague construction of “outrageously or wantonly vile, horrible, or inhuman” aggravating factor violated both the Eighth and Fourteenth Amendments.

Enmund v. Florida, 458 U.S. 782 (1982): Holding that a death sentence for defendant who aids and abets a felony in the course of which murder is committed by accomplices, but who does not himself kill or intend that killing take place violated the Eighth and Fourteenth Amendments. Professor Liebman himself represented defendant in this case.

Mills v. Maryland (1988): Holding that a statute which precludes a jury from considering any mitigating evidence unless it is unanimous violates the Eighth and Fourteenth Amendments.

The foregoing, while abbreviated, is a representative sample of watershed decisions that profoundly altered and reshaped the legal landscape well after 1973 and into the next decade. And, as observed by Professor Wilson, it is not at all clear from “A Broken System” what percentage of the reversals reported were in fact attributable to “these big changes in rules.” It may reasonably be assumed, however, that the percentage is not insignificant.

Other criticisms have been lodged against “A Broken System.” Nevada’s Attorney General recently took issue with Liebman’s methodology, noting that although that state’s death penalty records are kept by the Nevada Supreme Court, Attorney General, Department of Prisons, 17 district attorney and 17 court clerks, Liebman elected to obtain case information from criminal defense attorneys and the NAACP Capital Punishment Project, an organization committed to the abolition of the death penalty. Liebman stated that, in Nevada, there were 34 reversals in 108 cases, for an error rate of 38%. Conducting an independent analysis of error rate, the Nevada Attorney General’s Office, however, found that of 152 cases, only 30 death sentences were reversed. The actual error rate was thus a much lower 19%.

In addition, Nevada’s Attorney General’s Office echoed Professor Wilson’s concern that “A Broken Window” may propagate the plainly erroneous perception that the innocent are being unjustly executed. On the contrary, the Nevada Attorney General’s Office, in its assessment of death penalty cases, noted that most reversals were based on “attorney or judge procedural error,” and others occurred “where juries followed the existing law, but later the Supreme Court changed it.” At bottom, the Nevada Attorney General’s Office expressed the common-sense perspective that a 19-23% error rate in all death penalty cases tried since 1973 hardly demonstrates a system in disrepair. Quite the contrary, it proves that Nevada’s appellate courts function as an effective screen against potential miscarriages of justice with respect to the implementation of the death penalty.


At present, the reversal rate of New Jersey capital cases reviewed on direct appeal exceeds the overall 68% reversal rate touted by Professor Liebman in his study. Specifically, since the reinstitution of capital punishment in this State in 1982, the Supreme Court has, commencing with State v. Ramsuer, 106 N.J. 123 (1987), reviewed 51 death sentences on direct appeal. Of those 51 cases, the Court found reversible error in 36, or 70%, of all cases examined.1 (Exh. 3). In 19 out of these 36 cases, or 52%, the Supreme Court affirmed the defendant’s capital murder conviction but vacated his or her death sentence.

Of greater significance, instructional error accounted for an astounding 66% (24) of all reversals. Moreover, a substantial percentage of the foregoing errors were attributable to two profound yet entirely unforeseeable (from the trial court’s perspective) changes or interpretations of the death penalty statute. The first of these developments was the Legislature’s amendment of the death penalty statute in 1985 requiring the State to prove, in order to obtain a death sentence, that any applicable aggravating factor or factors outweigh beyond a reasonable doubt any applicable mitigating factors, and the Court’s subsequent decision in State v. Biegenwald (II), 106 N.J. 13 (1987) that the death penalty could not be imposed, irrespective of whether the case occurred before or after the adoption of the 1985 amendment, “without a finding that the aggravating factors outweighed the mitigating factors beyond a reasonable doubt.”2 This holding resulted in the reversal of numerous death verdicts obtained prior to the Court’s decision in Biegenwald (II) (Lodato, Koedatich, Zola, Hunt, (Marie) Moore, Davis, and Pitts).

Equally unanticipated was the Supreme Court’s 1988 holding in State v. Gerald, 113 N.J. 40 (1988) that as a matter of state constitutional law, only those who knowingly or purposely cause death — as opposed to those who purposely or knowingly cause serious bodily injury resulting in death — were eligible to receive the death penalty. Subsequent to Gerald, the Court decided a number of cases which were tried or pled before that decision. The vast majority of these appeals resulted in reversals (Coyle, Long, Pennington, Dixon, Clausell, Harvey, and Erazo). In State v. Davis, 116 N.J. 341 (1989) and State v. Jackson, 118 N.J. 484 (1990), the Supreme Court vacated guilty pleas to capital murder because in neither case did the defendant establish under Gerald whether he intended to knowingly or purposefully kill. All told, the Gerald decision precipitated 10 (27%) of the 36 reversals.3

Also in 1988, the United States Supreme Court in Mills v. Maryland,468 U.S. 367 (1988) announced that a sentencer must be permitted to consider all mitigating evidence, and therefore a state may not constitutionally require mitigating factors to be found unanimously before they can be used in the weighing process. In the wake of Mills, death sentences were reversed in State v. Bey (II), State v. Hightower (I), and State v. Dixon. The United States Court of Appeals for the Third Circuit’s decision in Humanik v. Beyer, 871 F.3d 342 (3d Cir.), cert. denied, 493 U.S. 812 (1989), which declared New Jersey’s diminished capacity statute unconstitutional, compelled reversals in State v. (Samuel) Moore and State v. Oglesby, two cases tried before Humanik was decided. In State v. Purnell, 126 N.J. 518 (1992) the Supreme Court reversed defendant’s death sentence and remanded for further proceedings, reasoning that the failure to charge felony-murder in a case where an aggravating factor was predicated on the commission of an underlying felony was reversible error.

Lastly, in State v. Brown, 138 N.J. 481 (1994) and State v. Mejia, 141 N.J. 475 (1995) the Supreme Court vacated death sentences based on the trial courts’ failure to instruct the jurors in those cases that they need not be unanimous with respect to the finding of death eligible “triggers,” i.e., “own conduct” and “intent to kill.”

With respect to reversals unrelated to instructional error, in only two appeals, State v. Perry, 124 N.J. 128 (1991) and State v. (Marie) Moore, 113 N.J. 239 (1988) did the Supreme Court conclude that insufficient evidence had been introduced to establish the death eligible status of the defendants, although neither Perry nor Moore were factually innocent of the homicides they were charged with committing. In State v. DiFrisco (I), 118 N.J. 253 (1990) the Court concluded that defendant’s death sentence had to be set aside for want of any extrinsic corroboration of his confession. Following a new penalty phase at which the jury reimposed the death penalty, his second death sentence was subsequently affirmed in DiFrisco (II).

Several capital defendants have raised claims of ineffective representation on direct appeal (Davis, Savage, Dixon, Marshall, Chew, and DiFrisco), yet only one, Savage, prevailed in overturning his conviction and death sentence on that basis. Notably, on remand the State obtained a capital murder conviction with regard to one of Savage’s two victims and a non-capital verdict of murder as to the other. At Savage’s second penalty trial, the jury unanimously found two aggravating factors but was unable to agree on the appropriate penalty. Obviously, Savage was anything but “innocent” of the killings. Furthermore, ineffective assistance of counsel claims have been rejected by the Supreme Court in all three appeals from the denial of post-conviction relief by those capital defendants, Marshall, Martini, and Bey, whose cases have proceeded to that stage of appellate review. There is thus no basis whatsoever to conclude that capital defendants — particularly indigent defendants represented by the Office of the Public Defender — have received and continue to receive anything less than exceptional representation at both the trial and appellate level.

In marked contrast to Professor Liebman’s finding of widespread prosecutorial misconduct predicated on the withholding of exculpatory material in other jurisdictions, only one death sentence in New Jersey has been vacated based on the withholding of alleged “Brady” material.4 In State v. Nelson, 155 N.J. 487 (1988) the Court overturned defendant’s death sentence for the murder of a police officer because the Camden County Prosecutor’s Office failed to turn over to the defense a complaint filed against it by the brother of one of defendant’s victims. The Court believed that evidence of the complaint would have been favorable to defendant in the penalty phase because “[t]he allegation that law enforcement personnel had been inadequately trained lent direct support to defendant’s catch-all mitigating factor.” The defendant’s plea to capital murder was left undisturbed by the Court. It bears noting that two justices strenuously disagreed with the majority’s finding of materiality, and expressed their view that the jury’s verdict of death was indeed “worthy of confidence.” In any event, the survey of capital appeals in New Jersey does not sustain Professor Liebman’s thesis that prosecutors nationwide all-too-frequently cast aside their oaths to do justice when pursuing capital verdicts.

Finally, in 13 cases reversed by the Supreme Court, prosecutors again attempted to secure capital verdicts. Five of those retrials resulted in the reimposition of the death penalty (Bey, Biegenwald, Hightower, Harvey, and DiFrisco).5 Prosecutors obtained murder convictions in the remaining cases. Indeed, with one exception, Walter Oglesby6 , every capital defendant whose death sentence was not ultimately affirmed by the Supreme Court presently stands convicted of murder. Put differently, no jury in this State has ever convicted and sentenced to death a factually innocent defendant.


Ultimately, New Jersey’s relatively high reversal rate is predominantly attributable to early yet profound refinements to this State’s death penalty statute wrought by the Supreme Court of New Jersey and the Legislature. Little else can explain why the Court is now strongly inclined to affirm, rather than reverse, death sentences on direct appeal.7 Indeed, in the last ten direct appeals taken by capital defendants, the Supreme Court has reversed only one death sentence (Nelson). This development contrasts sharply with the period prior to Marshall (I) when the Court reversed every death sentence it scrutinized in an unbroken succession of opinions. Moreover, the Court has yet to reverse a death sentence on grounds of disproportionality or reverse a trial court’s denial of post-conviction relief in a capital case.

Our Supreme Court is, furthermore, unrelenting in its effort to monitor for the presence of racism in the administration of the death penalty, as evidenced by its appointment of a standing special master, the Honorable David S. Baime, P.J.A.D., to conduct annual systemic studies of the death penalty. Indeed, this past term in In re Proportionality Review Project (II), 165 N.J. 206 (2000) the Court unambiguously embraced Judge Baime’s finding that no reliable evidence exists demonstrating that the race of the defendant or victim plays any role in the imposition of capital punishment in New Jersey. When all is said and done, reversal rates, as reflected by New Jersey’s experience, do not accurately measure whether the death penalty “works” as an effective system of punishment. Following a relatively brief period of refinement and clarification of the death penalty statute and death penalty practice, our Court is obviously confident that death penalty verdicts returned in New Jersey are exceptionally fair and just. Contrary to the portrait of the death penalty drawn by one of its harshest and prominent critics, Professor Liebman, New Jersey’s system of capital punishment has never ensnared the innocent, nor is it tainted by racism. At bottom, New Jersey’s death penalty works, and it works exceedingly well.

[1] Interestingly, Liebman’s findings confirmed a complaint by supporters of the death penalty who say the appeals process is entirely excessive. Liebman, however, hypothesizes that the reason capital sentences spend so much time awaiting judicial scrutiny is precisely because they are so persistently and systematically fraught with “alarming amounts” of error. A more compelling explanation for delay was articulated by Judge Alex Kozinski of the United States Court of Appeals for the Ninth Circuit. In a lecture on the death penalty reprinted in the Case Western Reserve Law Review, Judge Kozinski argued that “[t]he simple fact is the process takes so long because there is a concerted effort afoot to slow it down, and because our legal system requires scrupulous review before a death sentence can be carried out.” Judge Kozinski wryly observed that “[i]t is somewhat akin to the classic definition of chutzpah for death penalty opponents to say we can’t execute someone too fast because he is entitled to a searching review, and then to say what we are doing is immoral when we delay the execution precisely to afford such review.” The Honorable Alex Kozinski, Death: the Ultimate Run-On Sentence, 46 Case W. Res. L. Rev. 1, 25 (1995).

1 A trial court vacated Raymond Kise’s death sentence in 1987 because of error in the charge with respect to the jury’s consideration of aggravating factors. Kise was spared the death penalty at a subsequent penalty trial before a judge.

2 When enacted in 1982, the statute authorized the imposition of the death penalty if the aggravating factor or factors proven beyond a reasonable doubt were not outweighed by one or more mitigating factors.

3 In response to Gerald, the New Jersey Constitution was amended in 1992 by the voters to permit “SBI” murderers to be sentenced to death.

4 The Supreme Court rejected “Brady” claims alleged in Marshall (I) and (III), DiFrisco (I), and Martini (IV).

5 Biegenwald and Hightower’s second death sentences were subsequently reversed by the Supreme Court.

6 Walter Oglesby pled guilty to aggravated manslaughter in February 1992.

7 The trend toward affirmances is depicted in Exhibit 1.