Death by election?

A UT professor says voters have changed the way the Supreme Court decides death penalty cases

How Tennessee justices have voted in capital cases since Justice Penny White left the bench

By Daniel J. Foley

The outcome of death-penalty appeals in Tennessee has undergone two dramatic shifts in recent years, both of which can be traced to the ballot box.

The first, the 1990 election of three new justices, moved the five-member Tennessee Supreme Court in a liberal direction on capital punishment. The second, voter ouster of liberal Justice Penny J. White in 1996, precipitated a dramatic shift to the right.

This article, which analyzes the court's voting record on capital cases since the defeat of Justice White, is based on the author's database of Supreme Court decisions, which goes back to May 1989 and includes opinions filed through the end of the year 2000. For purposes of analysis, the court's recent history has been divided into three eras: the Old Court, the New Court and the Post-White Court.

Here is what the database shows:

• The Old Court - the five justices who served prior to the 1990 elections1 - wrote 12 opinions on capital punishment between May 1989 and the end of its term, and upheld the death penalty in 11, or 92 percent.
• The New Court - which included the three newly elected justices2 - wrote 27 opinions through 1996, and upheld the death penalty in 14, or 52 percent. Justice White was defeated in a yes-no retention election in 1996, after a campaign that focused on her concurring vote to overturn a death-penalty sentence.
• The Post-White Court - which has two new justices3 - wrote 30 opinions, involving 33 defendants, between 1997 and 2000. It upheld the death penalty for 30 of the 33 defendants, or 91 percent.

The above figures include only those cases on direct appeal, in which the court decided specifically on the merits of the death penalty or on proceeding with an execution. The figures do not include cases on post-conviction appeal, which will be discussed later in this article.

While there has been a dramatic ideological shift to the right on death-penalty cases in the Post-White era, there has not been any noteworthy change on the civil side. Both the New Court and the Post-White Court tended to vote in a liberal direction more often than conservative on civil matters. The overall ideology during the three court eras also will be considered later in this article.

But first, we'll turn our attention to death-penalty cases. The court has two new members since Justice White's ouster. Justice Janice M. Holder was appointed in December 1996 to fill the White vacancy and Justice William M. Barker was appointed in April 1998 after Justice Lyle Reid resigned. The two newest justices have been much more inclined than their predecessors to vote for the death penalty.

Here are the pro-death-penalty voting percentages for individual justices in the Post-White era:

• Justice E. Riley Anderson: 91 percent
• Justice Janice M. Holder: 91 percent
• Justice Frank F. Drowota III: 90 percent
• Justice William M. Barker: 84 percent
• Justice Lyle Reid: 50 percent
• Justice Adolpho A. Birch Jr.: 36 percent

The case-by-case votes of individual justices are listed in Table 1 (page 21). The only cases in which Justices Anderson, Holder, Drowota and Barker voted against the death penalty involved two in which the court reversed convictions of defendants who were sentenced to death and one in which it reversed the sentence.

In a Sept. 7, 1999, opinion by Justice Drowota, a unanimous court tossed out the premeditated first-degree murder conviction of defendant Pat Bondurant after concluding that the jury was not properly selected nor properly sequestered.4

In a Dec. 11, 2000, opinion, also by Justice Drowota, the court reversed the multiple premeditated murder convictions of defendant James Montgomery, ruling that Montgomery should have been tried separately from co-defendant Tony V. Carruthers. The court upheld the death sentence against Carruthers, with Justice Birch dissenting.5

In a Feb. 1, 1999, opinion by Justice Holder, the court upheld the felony murder conviction of defendant Preston Carter, but unanimously reversed the death sentence because the trial court judge used an outdated verdict form. That form did not include the required language stating that the prosecution had proved "beyond a reasonable doubt" that the statutory aggravating circumstances outweighed any mitigating circumstances.6

Other than those three cases, the Post-White court has consistently affirmed jury sentences in death-penalty cases. And, perhaps most significantly, in a March 6, 2000, opinion by Justice Drowota, the court affirmed a trial court finding that death-row inmate Robert Glen Coe was competent to be executed, shortly before Coe became the first prisoner in 40 years to be put to death in Tennessee. Justice Birch dissented.7

Justice Drowota has written more than half of the court's majority opinions on death-penalty cases in the Post-White era - 17 of 30. Justice Anderson wrote seven and Justices Holder and Barker three each.

Justice Birch's dissenting opinions

Justice Birch did not write any majority opinions, and he has been the lone dissenter on capital punishment cases since the departure of Justice Reid. Overall, Justice Birch voted against the death penalty for 21 of the 33 defendants whose cases were considered in the Post-White era. In addition to the three cases in which he joined the majority in overturning a conviction or death sentence, Birch cited the following reasons for opposing the death penalty:

• Inadequate review of the proportionality of the death sentence (seven times).
• Insufficient evidence of, or improper application of, the death-penalty law relating to aggravating circumstances (three).
• Admission of prejudicial victim impact evidence (two).
• Improper jury instructions (two).
• Insufficient evidence the murder was premeditated.
• Extenuating circumstances justifying commutation of a death sentence.
• Inadequate method of evaluating a defendant's competency to be executed.
• Insufficient time for defendant to appeal an order of execution.

Both Justices Reid and Birch were particularly troubled by what they saw as the court's inadequacies in its "proportionality review" of death sentences. The purpose of proportionality review is to evaluate in each case whether capital punishment is excessive when compared to similar cases, considering the nature of the crime and the defendant. In general, the purpose is to limit death sentences to the "worst of the bad."

The court majority, in a 1997 opinion by Justice Drowota in State v. Bland,8 attempted to articulate the standards for proportionality review: "If the case, taken as a whole, is plainly lacking in circumstances consistent with those in similar cases in which the death penalty has been imposed, the sentence of death in the case being reviewed is disproportionate."9

The court said that each death-penalty sentence should be compared to a "pool" of all previous cases in which a capital sentence hearing was conducted to determine whether the defendant should be sentenced to life in prison, life without the possibility of parole or death.10

In reviewing previous court decisions for relevant factors, the court said the following are among those that should be considered for purposes of comparing the nature of the crime: (1) means of death, (2) manner of death (violent, torturous, etc.), (3) motivation for killing, (4) place of death, (5) similarity of victims' circumstances including age, physical and mental condition, and treatment during killing, (6) presence or absence of premeditation, (7) presence or absence of provocation, (8) presence or absence of justification, (9) injury to and effects on victims who were not killed.11

The court said the following factors should be among those considered in regard to the defendant's characteristics: (1) prior criminal activity, (2) age, race and gender, (3) mental, emotional and physical condition, (4) role in the murder, (5) cooperation with authorities, (6) remorse, (7) knowledge of the helplessness of victims, (8) capacity for rehabilitation.12

Justice Reid concurred with the majority in upholding the conviction of Andre S. Bland, but dissented on the death sentence, which he said was disproportionate to previous cases involving similar circumstances. Reid conceded that the majority opinion was at least a good beginning in setting some standards for proportionality review. But, he added, the court had reviewed 116 cases for proportionality and never found one that was excessive, so it remains to be seen whether the procedure will produce "more than the routine affirmation of jury verdicts."13

In a separate concurring and dissenting opinion, Justice Birch agreed with Reid and cited an additional concern about proportionality review - the relatively small number of cases used for comparison. Rather than limiting the "pool" to cases in which the state sought the death penalty, Birch said, the court should compare each death sentence to the sentence in all first-degree murder convictions.14

Justice Birch expanded on his concerns regarding proportionality review in a concurring and dissenting opinion in State v. Chalmers, by suggesting there is evidence that the "pool" of comparison cases in Tennessee may be "race-tainted" - meaning racial prejudice influenced either the prosecutor's decision to seek the death penalty or the jury's decision to impose it.15

"If a defendant's sentence is compared to a pool of cases which are similarly race-tainted, the reviewing court is without a benchmark by which to determine whether the defendant's sentence is "aberrant,"" he said. "Furthermore, the majority does not clarify whose race should be considered in comparative proportionality review - the defendant's, the victim's, or both ... "16

"Given the vast array of evidence indicating that racial disparities in death sentencing are at their greatest in cases where an Afro-American defendant has killed a Caucasian person, we should clearly establish that the victim's race and the defendant's race bear equal significance in comparative proportionality review."17

In summing up his concerns about proportionality review, Justice Birch said: "The review procedures are ineffective for three reasons: the "test" we employ is so broad that nearly any sentence could be found proportionate; our review procedures are too subjective; and the "pool" of cases which are reviewed for proportionality is too small."18

Echoing Reid's skepticism of the court's approach to proportionality review, Birch noted: "In the 14 capital cases the Supreme Court has reviewed for proportionality since Bland, the string of affirmations remains unbroken by even a single reversal."19

Justice Birch also voted against the death penalty in several cases after concluding that the admission of victim impact evidence was prejudicial. In State v. Nesbit, Birch wrote: "Generally, victim impact evidence is unsettling because its use encourages the jury to quantify the value of the victim's life and urges the finding that murder is more reprehensible if the victim is survived by a bereaved family than if a victim had no family at all." 20

In a dissenting opinion in State v. Smith,21 Justice Birch went a step further and said he would overrule court precedent on the admissibility of victim impact evidence to the extent that the controlling case, State v. Payne,22 permits unlimited or vaguely limited victim impact evidence.

"I would hold that the admission of victim impact evidence during the sentencing phase of a capital punishment trial is unconstitutional under Tennessee law unless admitted with precise definition and clear limitation," he said.23

Birch cited a New Jersey procedure as a model for reducing the possibility that jurors will misuse victim impact evidence. The New Jersey procedure requires that the defendant be notified in advance if the prosecution intends to introduce victim impact evidence, that the defendant's attorney be given the opportunity to interview victim impact witnesses prior to their testimony, that testimony generally be limited to one survivor and that minors not be permitted to present evidence unless there are no suitable adult survivors.

The New Jersey procedure also requires the trial court to conduct a hearing outside the presence of the jury to determine the admissibility of victim impact evidence and requires the victim impact witness to reduce testimony to writing and to limit testimony to factual, not emotional matters. A witness is not allowed to testify if the witness cannot control his or her emotions and the witness is permitted only to read the previously approved testimony.24

Protocols in capital appeals

In addition to ruling on the merits of death sentences in individual cases, the Post-White court has utilized its death-penalty opinions to establish standards or protocols of review on several legal issues arising in capital cases.

One of those, dealing with proportionality review of a specific death sentence, was set forth in State v. Bland, and was discussed above.

In another case, State v. Reid, the court addressed the question of how a defendant may introduce evidence of his or her mental condition as mitigation in the sentencing phase of a capital trial. In a unanimous opinion by Justice Drowota, the court established these standards: (1) a defendant must file pre-trial notice of intent to present expert testimony on mental condition, (2) the trial court may, at the request of the prosecution, order the defendant to undergo psychiatric evaluation by a mental health expert selected by the state, and (3) the defense will have access to any expert reports prior to the trial, while the state will have access only after a verdict of guilty and the defendant's confirmation of his or her intent to offer expert evidence on mental condition.25

Such a procedure would not violate the defendant's protection against self-incrimination, the court said, adding: "Requiring a capital defendant to provide pretrial notice of intent to introduce expert mitigation proof relating to mental condition protects the state's right of rebuttal and eliminates unjustifiable delay."26

In another case in which the court crafted a procedural set of standards where none existed, the justices addressed the question of when, where and how a convicted murderer may raise the issue of competency to be executed. Justice Drowota wrote the majority opinion in Heck v. State, which held that in Tennessee a prisoner is incompetent to be executed if the he or she "lacks the mental capacity to understand the fact of the impending execution and the reason for it."27

The court declined to include in the definition of incompetency a defendant's inability to assist counsel: "Once the conviction is final, there is a lessened need for a defendant to assist in his or her defense given the availability of both state and federal collateral review of trial errors, and the expansion of the right to competent counsel at trial."28

A prisoner may not raise the issue of competency until execution is "imminent," meaning all state and federal appeals regarding conviction and sentence have been exhausted and the state Supreme Court has set an execution date upon motion by the attorney general, the court said.29

The place for raising the issue of competency is before the trial court where the defendant was tried and sentenced to death, the court added, and the burden is on the prisoner to make a threshold showing that his or her competency is genuinely at issue. If that burden is met, the prisoner then has the burden of overcoming the presumption of competency, and must do so by a preponderance of the evidence.30

In a concurring and dissenting opinion, Justice Birch criticized three aspects of the protocol set forth by the majority. The criteria for competency should include an inquiry into whether the prisoner has the ability to assist counsel, the decision on competency should be made by a jury rather than a judge and, once the prisoner has made a threshold showing of incompetence, the burden of proof should be on the state, he said.31

Post-conviction appeals

In addition to direct appeals on death-penalty cases, the Tennessee Supreme Court occasionally considers post-conviction appeals in which defendants who lost on direct appeal argue for relief by claiming denial of a constitutional right.

The Post-White court decided nine cases on post-conviction appeal and denied relief in all, including one in which the court reversed a Court of Appeals decision favoring the defendant.32 The total of nine cases includes one - Heck v. State, described above - in which the court decided that post-conviction appeal was not an appropriate avenue for litigating the issue of competency to be executed.

In four of the post-conviction appeals, prisoners raised issues about the interpretation of the state's felony murder law, two challenged the interpretation of the law relating to aggravating circumstances in death-penalty cases, two argued they were denied effective assistance of counsel and one raised the issue of competency to be executed. Although the court found errors in four of the nine cases, the judges ruled that those errors were harmless and did not affect imposition of the death sentence.

The case-by-case votes of individual justices on death-penalty post-conviction appeals are listed in Table 2 (page 22). Here is how they voted (votes to deny relief, in effect to uphold the death penalty, are listed first, followed by votes to grant relief): Drowota 9-0, Holder 9-0, Barker 5-0, Anderson 8-1, Birch 7-2, Reid 1-2.

If all of the death-penalty opinions written by the Post-White court are added up - both from direct and post-conviction appeals - the court upheld the death penalty against 39 defendants and reversed the penalty against three.

Overall ideology

The court's shifting ideology over the three eras - from conservative to liberal and back to conservative - can be traced primarily to the justices' opinions on capital punishment and other criminal matters. Table 3 (page 23) shows the court's leanings based on votes in all cases that were suitable for ideological analysis.33 The Old Court decided cases in a liberal direction 43 percent of the time, the New Court in 56 percent and the Post-White Court in 40 percent.

When criminal and civil cases are separately analyzed, however, an interesting pattern emerges. While court opinions favored the prosecution over the defendant more than half of the time across all three eras, differences among eras are dramatic: Here are the liberal voting percentages: Old Court 19 percent; New Court 47 percent; Post-White Court 23 percent.

By contrast, the court voted in a liberal direction in a majority of civil cases through all three eras, and the differences among eras are not statistically significant. The liberal voting percentages were: Old Court 52 percent; New Court 62 percent; Post-White Court 60 percent.

The ideological shift from the liberal leanings of the New Court to the conservative leanings of the Post-White Court can be attributed to at least two factors. First, and most obvious, two justices with generally liberal leanings were replaced by justices with more conservative views. Justice White, who had a 60-percent liberal voting record, was replaced by Justice Janice M. Holder, who has a 37-percent liberal record. Justice Lyle Reid, who was on the court during the New Court era and into the Post-White era until resigning in 1998, voted liberal 58 percent of the time. His replacement, Justice William M. Barker, has voted liberal 42 percent of the time.

A second, and perhaps less obvious, reason is that the holdover justices are now showing more conservative voting tendencies. Table 4 (page 23) shows that Justice E. Riley Anderson had a 55-percent liberal voting record in the New Court era, and a 41-percent liberal record in the Post-White era. Justice Frank F. Drowota III went from 53 percent liberal to 37 percent. Justice Adolpho A. Birch Jr. went from 54 percent to 47 percent.34 Even Justice Reid went from 61 percent liberal in the New Court era to 49 percent liberal in the Post-White era. Readers may want to speculate about the applicability of Mr. Dooley's observation that judges follow the election returns.35

An analysis of the database suggests a couple of factors that may help explain the shift in voting patterns of individual justices. One factor is a change in the mix of cases. The court is now hearing fewer worker compensation cases than it once did, thus reducing the number of cases that generally have a liberal outcome. At the same time the court is now hearing more criminal cases than it once did, thus increasing the number of cases that generally have a conservative outcome.

A second factor is the degree of unanimity on the court. Although the number of split votes - those in which at least one judge wrote a dissenting opinion - has increased from 6 percent to 11 percent to 15 percent over the three eras, the court nevertheless seems to strive for unanimity as much as possible. On close calls, that may mean that a justice who would be inclined to vote in a liberal direction if that is the consensus, may nevertheless choose not to write a dissenting opinion if the consensus is in a conservative direction.

Whatever the reasons, the Post-White court has a much more conservative voting record than the New Court, especially on criminal cases and most especially on death-penalty sentences.

Notes

1. The justices on the Old Court were: Williams H.D. Fones, Robert E. Cooper, William J. Harbison, Frank F. Drowota III, and Charles H. O'Brien.
2. The justices on the New Court were: Martha Craig Daughtrey (subsequently replaced by Adolpho A. Birch Jr.), E. Riley Anderson, Lyle Reid, Drowota, and O'Brien (subsequently replaced by Penny J. White).
3. The justices on the Post-White Court are: Anderson, Drowota, Birch, Reid (subsequently replaced by William M. Barker) and Janice M. Holder.
4. State v. Bondurant, 4 S.W.3d 662 (Tenn. 1999).
5. State v. Carruthers, 35 S.W.3d 516 (Tenn. 2000).
6. State v. Carter, 988 S.W.2d 145, 152 (Tenn. 1999).
7. Coe v. State, 17 S.W.3d 193 (Tenn. 2000).
8. 958 S.W.2d 651 (Tenn. 1997).
9. Id. at 665.
10. Id. at 666.
11. Id. at 667.
12. Id. at 667.
13. Id. at 675-676.
14. Id. at 679.
15. 28 S.W.3d 913, 921 (Tenn. 2000).
16. Id. at 923.
17. Id.
18. Id.
19. Id. at 925
20. 978 S.W.2d 872, 903 (Tenn. 1998), citing Booth v. Maryland, 482 U.S. 496 (1987)
21. 993 S.W.2d 6, 22 (Tenn. 1999).
22. 791 S.W.2d 10 (Tenn. 1990).
23. Smith, at 26.
24. Id. at 25.
25. 981 S.W.2d 166, 168 (Tenn. 1998).
26. Id. at 172.
27. 6 S.W.3d 257, 266 (Tenn. 1999)
28. Id. at 266.
29. Id. at 267.
30. Id. at 267-271.
31. Id. at 275.
32. Henley v. State, 960 S.W.2d 572 (Tenn. 1997).
33. The ideological direction of each opinion was determined by which party prevailed. In a criminal case, a decision was coded as liberal if it favored the defendant and conservative if it favored the prosecutor. In a torts/damages case, a decision was liberal if it favored the injury claimant and conservative if it favored the defendant or insurance company. In a worker compensation case, a decision was liberal if it favored the worker and conservative if it favored the employer. In general, an opinion was classified as liberal if the party that prevailed was the one perceived as having less economic or social influence and conservative if the opposite was the case. When a case involved adversaries who were not readily identifiable by ideological classification - such as a divorce case - it was excluded from the analysis. So were cases with no clear-cut winner.
34. The difference in Justice Birch's voting record in the New Court and Post-White eras is not statistically significant. The differences for the other holdover justices are statistically significant.
35. Mr. Dooley was created by American journalist and humorist Finley Peter Dunne, 1867-1936. Among Mr. Dooley's oft-quoted sage sayings was this one: "No matter whether the constitution follows the flag or not, the supreme court follows the illiction returns."

_____________________

Daniel J. Foley is an associate professor of journalism at the University of Tennessee, Knoxville. One of his areas of research interest involves use of databases and statistics to analyze court decisions. He is the author of Tennessee Supreme Court: A Statistical Analysis of an Ideological Shift After the 1990 Election, 64 Tenn. L. Rev. 155-193 (1996). He can be reached at dan-foley@utk.edu. Research for this article was supported by a grant from the College of Communications, University of Tennessee.