A comparison of the criminal appellate decisions of appointed state supreme courts: insights, questions, and implications for judicial independence.

Of all the systems used to select judges in the United States, appointment systems are the most widely used. (1) An appointment system is one in which the state's governor, with or without the input of a nominating commission, chooses candidates to fill initial and interim vacancies on a court. (2) Today, a majority of the states use appointment systems to select their supreme court judges. (3) There are many important differences in the institutional arrangements and procedures that these appointment systems use, however, and no research has been done to see if these differences affect the outcome of cases.

This Article presents the results of a study conducted to see whether state supreme courts selected in states with dissimilar appointment systems differ in the way they decide criminal appeals. Comparing the criminal decisions of courts selected with different appointment systems may also suggest something about how different appointment systems impact judicial independence.

I. APPOINTMENT: THE FALL AND RISE OF A JUDICIAL SELECTION SYSTEM

As of 1846, appointment was second the most common way of selecting justices of the states' highest courts. (4) Of the twenty-nine states that entered the union prior to that date, fourteen used appointment systems to select supreme court justices. (5) Over the course of the nineteenth century, however, the proportion of states that used appointment systems for their highest courts dropped sharply as many new states entered the union with different selection systems (typically partisan elections) and as states already in the union switched to different selection methods. (6) The change away from appointment toward partisan election of justices was, among other things, prompted by the belief that elected judges would exercise their duties more independently than judges who owed their appointments to the governor or to the legislature. (7) The change was also spurred by the belief that elections would prevent the judiciary from being filled with judges who owed their appointments to political connections rather than to personal qualifications. (8) Defenders of judicial selection systems that combined appointment and life tenure retorted that elections would undermine judicial independence by subjecting judges to the will of the people and to manipulation by political party leaders; but these arguments did not prevent the adoption of judicial elections in most states during the period. (9)

By 1909, thirty-five of the forty-six states in the union used partisan elections to select supreme court justices. Partisan elections, which some viewed as a means of assuring judicial independence from the other branches of government, came under renewed criticism for being detrimental to judicial independence. Critics claimed that elected judges were indebted to the political parties upon whom the judges depended for electoral support and that party leaders could use their patronage powers to influence the judiciary. (10) Critics also pointed out that elected judges frequently heard cases involving attorneys and litigants who had contributed to their campaigns, which further undermined public confidence in the courts' independence. (11)

Interest in appointment as a judicial selection method enjoyed a revival during the 1910s in response to the previously-mentioned problems. (12) Prominent proponents of a return to appointment during this period included Roscoe Pound, John Wigmore, and Albert Kales. (13) Furthermore, the effort to encourage judicial selection reform provided one of the driving forces behind the creation of the American Judicature Society in 1913. (14) In 1914, Kales proposed an appointment plan eventually known as merit selection, which many reformers of the era supported. (15) Kales's idea sparked the revival of interest in appointment as a judicial selection system. Although various states considered several versions of Kales's proposal throughout the 1930s, Missouri was the first state to adopt a merit selection plan in 1940. (16) Currently, twenty-three states and the District of Columbia use merit selection commissions to select the judges of their highest courts. (17)

As a system for choosing judges, merit selection attempts to accomplish three important goals: 1) enhance the professionalism of the judiciary, 2) enhance the independence of the judiciary, and 3) minimize the influence of partisan politics on the judicial selection process. (18) Under a merit selection plan, candidates for a vacancy on the bench are vetted by a commission (19) composed of lawyers, non-lawyers, and sometimes sitting judges. (20) After considering all of the applicants, the commission sends a list of the candidates that it deems most highly qualified (generally two to seven) (21) to the governor, from which the governor selects one candidate to fill a vacancy. (22) For proponents of merit selection, the use of commissions to vet candidates for judicial vacancies provides the primary advantage over other judicial selection systems because the focus of the candidate search remains on the professional and personal qualifications of the candidates, while minimizing political considerations in the selection process. (23)

Merit selection plans are frequently combined with retention elections as a means of re-selecting judges after their initial terms of appointment expire. Although forcing judges to run in any type of election introduces the possibility that they will be subject to influence by campaign contributors, (24) retention elections coupled with merit selection ensures that judges remain accountable to the people they serve. (25) In a retention election, the incumbent judge does not run against an opponent. (26) Instead, voters decide whether the judge should serve another term or be removed from office. (27) Despite the fact that judges do not face opponents on the ballot in retention elections, these elections can turn into expensive and bitter contests when special interest groups mount campaigns for a judge's removal. (28)

Between 1940 and 1989, twenty states (29) adopted appointment systems for selecting their supreme court justices, with fourteen of them switching from partisan or non-partisan elections to an appointment system using a merit selection commission and retention elections. (30) The fact that so many states changed from elections to appointment reflects the success of proponents of merit selection in making their case that merit selection, and appointment systems more generally, result in a more professional and independent judiciary.

Still, substantial variation remains in how appointment systems work in different states for filling supreme court vacancies. For example, California, Maine, New Hampshire, and New Jersey allow the governor to make both initial and interim appointments to their supreme courts without the use of a commission. (31) Furthermore, among the states that use commissions, (32) tremendous variation exists in the way that the commission members are chosen and how they do their work. For example, in some states the governor appoints all of the commission members, whereas in other states the state bar, the state legislature, or the judiciary plays a role in choosing the members. (33) States that use a commission to vet applicants for judicial vacancies use procedures that vary over whether the governor must choose a candidate from the list presented by the commission, (34) whether the legislature confirms the selected candidate, (35) and in the procedures used and factors considered by the commissions when reviewing the candidates. (36) States that use appointment to initially select their high court judges also vary as to whether the justices face reselection. (37) When justices face reselection, states also differ on the reselection process (38) and how long judges' terms of office last. (39) For example, most states that use merit selection commissions to select the judges of their high courts subject them to retention elections. (40) Some states, like Massachusetts (41) and Rhode Island, however, do not subject their high court judges to any form of reselection once appointed and confirmed. Other states, like Hawaii, require supreme court judges to go through the commission process again when their ten-year terms of office expire. (42)

II. THE POLITICAL SALIENCE OF CRIMINAL DECISIONS IN THE CURRENT STUDY

Although some researchers investigated the ways in which different selection systems affect how judges decide cases, most of this research has dealt with the differences between appointed judges and elected judges. (43) As stated earlier, there has been no research into how courts selected with dissimilar appointment systems differ in their decisional behavior. For example, no studies compare the rulings of a court appointed without a commission with the rulings of a court appointed with the help of a commission. Nor has research been conducted to see whether the manner in which judicial appointment systems function impacts how courts decide cases.

Research into the effects of different appointment systems on court decisions is crucial to understand how the various types of appointment systems accomplish one of their primary goals: the promotion of judicial independence. (44) The term judicial independence refers to the ability of judges to render decisions free from political or popular influence. (45) Of course, no indisputable method exists to scientifically measure a court's independence or the influence by other branches of government and public or private pressure on a court's decisions. One measurable manifestation of a court's independence might be its propensity to make decisions contrary to the political interests of the judges who comprise it. If a court's judges are not concerned with the opinions of those who could remove them from office, the hallmark of an independent court, arguably that court should be more prone to issue unpopular decisions, assuming all other factors are equal.

Accordingly, this study focuses on state supreme court decisions in criminal cases. There is anecdotal and empirical evidence which demonstrate that judicial decisions in criminal cases significantly impact the removal of judges from the bench. Although many other factors can lead to the removal of a judge, a reputation for being "soft" on criminal defendants carries well-documented negative repercussions. For example, case studies indicate that a judicial reputation as pro-criminal defendant played a prominent role in the defeats of three supreme court justices in California in 1986, (46) and of supreme court justices in Tennessee and Nebraska in 1996 in retention elections. (47) Other researchers noted the extreme rhetoric that judicial candidates use in their electoral campaigns in order to establish their anti-crime credentials. (48) Furthermore, judges who face reappointment by political officials can also be in danger if they develop a reputation for being soft on crime. (49) Such a situation occurred in New York, when a judge on the state's intermediate appellate court resigned after the governor informed the judge that he would not be reappointed because of his pattern of ruling in support of criminal defendants. (50)

A study of televised campaign advertisements run by supreme court candidates in four states in 2000 illustrates the perceived importance of criminal justice issues to winning or maintaining judicial office. (51) This study showed that candidates devoted over fifty-two percent of their commercials to criminal justice issues, while they devoted only twenty-four percent of their commercials to civil justice issues. (52) Judicial candidates understandably tailor their campaigns to emphasize their anti-crime views or credentials given that another study found an inverse relationship between the votes received by an incumbent state supreme court candidate and the state's murder rate the previous year. (53)

Some studies also suggest that judges' desires to avoid being labeled as pro-criminal defendant influence their decisions in criminal cases. A study of trial judges in Pennsylvania found that the judges tended to mete out harsher sentences to criminal defendants as election day grew closer. (54) Interestingly, the Pennsylvania judges behaved strategically to diminish the possibility of defeat, despite the fact that incumbent Pennsylvania judges run in retention elections at the end of each term (55) in which judges are very rarely rejected. (56) This finding from the Pennsylvania trial courts confirms the results of an earlier study involving selected supreme court justices in Kentucky, Louisiana, North Carolina, and Texas, with a record of supporting the claims of criminal defendants in a majority of non-unanimous decisions. (57) This study showed that within two years of the next election, judges with generally procriminal defendant records were more likely to vote with the majority in upholding death sentences or the underlying conviction predicating the death sentence. (58)

III. RESEARCH METHODOLOGY

The prior studies suggest that judges' decisions in criminal cases can play a pivotal role in determining retention when their terms expire. Arguably, many judges perceive that being hard on criminals is the "safest" political course of action to assure re-selection. Accordingly, examining rulings in criminal cases might suggest something about how public opinion, pressure from special interest groups, and the opinions of political officials who control their reselection influence judges.

Clearly, any inferences about judicial independence based on the tendency of a judge to rule in favor of criminal defendants must be made with care. I do not imply that a completely independent judge would never rule in favor of the government in criminal cases. Judges never compile perfectly anti-government records in criminal cases because many different factors influence decisions in these cases. These factors include case facts, (59) victim characteristics, (60) perpetrator characteristics, (61) judges' ideological preferences, (62) and the political ideology of the state. (63) Because of the incentives that exist for judges to appear tough on criminals, however, any tendency to back the claims of the accused in criminal cases may provide a rough and indirect indication of judicial freedom to make decisions that are politically unpopular.

A. How the Supreme Court Decisions in Criminal Cases Were Counted

Although the discussion up to this point focused on the motivation of individual judges to appear tough on crime, this study assumes that when judges vote together as a court, each judge retains an incentive to appear tough on criminals in order to secure reselection. Of course, different judges on a court might tend to favor the state more often in criminal cases in some circumstances than others, such as when a judge's term is nearing its end. (64) This study assumes that the desire to secure reselection creates an incentive for all judges to appear tough on crime. Furthermore, focusing on the outcome of cases, as opposed to focusing on the votes of individual judges, illustrates whether differences in appointment systems actually affect the fate of criminally accused and convicted persons who come before the states' highest courts.

The study examines cases decided by supreme courts in Alaska, Colorado, Connecticut, Hawaii, Maine, Massachusetts, New Jersey, Rhode Island, Tennessee, and Utah between 2000 and 2003. These states represent the geographical and political diversity of the country (65) and the state appointment systems used to select their high court judges differ in important ways. (66) This study examines state supreme court decisions in criminal cases brought on direct appeal, or by way of habeas corpus, and includes interlocutory appeals and final decisions in published cases. (67) When courts consolidated multiple cases, the outcome for each case in the group was counted separately because each of the cases possibly had a different outcome. This study excludes decisions on whether to grant leave to appeal, motions for reconsideration or rehearing, and unpublished decisions. Unpublished cases were excluded because of the difficulty of accessing such cases, and because they lack precedential value in all jurisdictions. (68) Furthermore, published decisions present, arguably, the most legally and politically influential products of appellate courts. (69) The study omitted cases in which lower court judges participated in judicial deliberations because these judges are sometimes chosen by different selection commissions, (70) and sometimes serve different terms of office from supreme court judges in the same state. (71) Finally, the study omitted cases in which retired supreme court judges sat by assignment because those judges no longer face reselection and might not be as concerned with relations between the court and the other branches of state government or public opinion as justices actively serving on the court.

Note also that a case was coded as being won by the state if the criminally accused or convicted party in the case was not granted the primary relief sought. For example, if a person appealed her convictions for auto theft and murder, and only the auto theft conviction was reversed on appeal, the case was coded as a state win because she failed to get all of the convictions overturned.

B. Significance of Selected Features of Appointment Systems

The study looks at how the following features of appointment systems are associated with the propensity of courts to decide cases in favor of the state: 1) whether a court's justices are selected using a merit selection commission, 2) whether a state subjects its supreme court judges to reselection, and 3) the length of the terms served by justices on a supreme court.

An association between one feature of an appointment system and court decisions in criminal cases can be found by taking the mean percentage of criminal cases decided in favor of the state by a group of courts with the feature in question and comparing it with the mean percentage of criminal cases decided in favor of the state by a group of courts without the feature. The difference in the mean percentages is then tested for statistical significance. (72)

C. Limitations of the Study

It is important to note some of the limitations of this study before explaining the results. Because the study does not include unpublished decisions, or decisions about whether to review cases from a lower court, it is not a comprehensive examination of the features of appointment systems associated with courts' decisional propensities in every type of decision that courts make. (73) Furthermore, this study does not examine the factors that influence which candidate a governor chooses to nominate, and it does not control for a host of issues that could affect the outcome of cases, such as the ideology of the justices who decide a case and the facts and the law at issue in the case. This inquiry merely compares the decisions of state supreme courts when certain appointment system features are present, and when they are not. The goal is to see if any systematic differences in the outcomes of criminal cases exist based on how a court's judges were chosen and retained. This analytical technique provides some clues as to whether these features impact court decision-making.

As indicated earlier, no research analyzed which features of appointment systems are associated with which decisional propensities among state supreme courts. A number of factors suggest expectations as to what the results of this study might show. For example, if supporters of merit selection commissions are correct that such commissions select more independent and professional judicial candidates, (74) then courts whose justices were appointed using merit commissions should decide fewer criminal cases in favor of the state on average than courts whose members were not selected with commissions. Furthermore, given that prior research suggested that judges are increasingly likely to rule in favor of the state in criminal cases as the end of their terms of office approach, (75) one would expect that courts with judges who serve longer terms and are not subject to reselection should rule in favor of the state less often than courts in states where the justices serve shorter terms or are subject to reselection.

IV. RESULTS

Table One presents a cross tabulation showing the percentage of criminal cases in the study that each court decided for the state and against the state:

Table Two presents the results of difference-of-means tests, comparing the rulings of courts in criminal cases to ascertain whether any differences in their rulings were associated with merit selection, term length, or the lack of reselection:

Overall, the results were startling and counterintuitive. Specifically, the results show that courts whose members were selected with the use of a merit selection commission decided a higher percentage of criminal cases in favor of the state than did courts chosen without the use of commissions. (76) The results also show that supreme courts whose justices do not face any sort of reselection decided a higher percentage of criminal cases in favor of the state than courts whose justices have to be reselected periodically. (77) Surprisingly, no statistically significant difference existed between the percentage of criminal cases decided in favor of the state by courts whose justices serve ten-year terms of office and courts whose justices serve shorter terms. (78)

V. DISCUSSION

These findings provide a number of interesting insights and raise a host of questions. The results regarding merit selection commissions suggest that there could be something happening in the selection process that causes the commissions to pick candidates who are more favorably disposed to the state in criminal cases and, therefore, more likely to make politically safe and popular decisions against criminally accused and convicted litigants.

The results in Table Three present more detailed findings regarding the impact of merit selection commissions on court decisionmaking in the presence or absence of reselection. Table Three shows that the use of reselection in conjunction with merit selection might mitigate the tendency to rule in favor of the state that we see in judicial selection systems using merit commissions. Table Three also shows that when one focuses only on states that use some form of reselection, the use of a merit selection commission is associated with a higher proportion of decisions in favor of the state.

The commissions' tendency to exclude more candidates whom they feel would not be sufficiently tough on criminals during the selection process could be facilitated by the fact that most commissions are required to keep their deliberations, communications, interviews, records, and votes on the candidates strictly confidential. (79) Some committees are even prohibited from revealing to the public the identity of the candidates under consideration. (80)

There is relatively little information available about the deliberations of judicial selection commissions, however, the data that exist possibly reveal something about how the bias of commissions against candidates who seem to be sympathetic to criminal defendants might manifest itself in commissions' deliberations. For example, in a survey of judicial selection commissioners in the 1970s, forty-nine percent of the respondents said that "political considerations" came up in their evaluations of judicial candidates. (81) The survey also asked if one quality would set a candidate apart from equally qualified peers. Of the 252 responses received by researchers, respondents mentioned "integrity" most frequently, while "judicial temperament" was tied for fourth, and "fairness" ranked seventh. (82) "Independence" tied for twentieth, falling below "a likeable personality," which tied for tenth. (83) Other studies of judicial selection commissions also reveal that political considerations play a significant and immeasurable role in the candidate evaluation process. (84) Still, more research is needed into the nature of the candidate evaluation process in which commissions engage, and into the factors that influence governors when they decide which candidates to nominate.

It is also unclear why states that do not subject their supreme court justices to any type of reselection have courts that are friendlier to the state in criminal appeals than courts in states where the supreme court justices are subject to reselection. The results in Table Four show that the proportion of cases that a state supreme court decides for the state appears to be lower if its judges are subject to reselection, regardless of whether the reselection is by reappointment or retention election, than if its judges are not subject to reselection. Table Four also shows, however, that there is no statistically significant difference in the proportion of cases decided in favor of the state between courts in states that use retention elections and courts in states that use reappointment.

Protecting judges from the fear of losing their jobs because of unpopular decisions has been touted as a way of enhancing their independence. (85) The results of this study suggest, however, that courts enjoying security of tenure appear more reticent to make potentially unpopular decisions by siding with criminal defendants than courts that do not enjoy such security. This finding clearly calls for more research into why security of tenure is associated with courts that make fewer politically unpopular decisions by siding with the state, but one possibility suggests itself. Courts may make decisions based on a host of reasons, including a desire to maintain the respect and esteem of friends, colleagues, the public, and the other branches of government. (86) Accordingly, these judges may be trying to protect their court budgets from cuts (87) or to protect the court from legislators diminishing the court's jurisdiction because of the legislators' displeasure with decisions favoring the criminally accused. (88) Also, researchers note the tendency of the United States Supreme Court to match the public's policy preferences with its rulings, despite the fact that its justices have life tenure, so that its rulings are not legislatively overturned or left unenforced. (89) Perhaps the same motives might prompt state supreme courts whose judges do not face tenure review to closely track public opinion with their rulings.

Public opinion and threats from the other branches of government likewise exert pressure on courts that are subject to reselection, and more research is needed to find out why these pressures possibly have a greater impact on courts that are not subject to reselection. One explanation could be that reselection gives courts an institutional legitimacy that encourages them to act more independently than courts whose judges do not serve with the direct endorsement of the state's voters, or who have to be reappointed by an elected governor. Supporters of judicial election advanced this argument regarding elections in the nineteenth century, (90) and Justice Larry Starcher of the West Virginia Supreme Court more recently advanced the same argument. (91) Furthermore, Justice Shirley Abrahamson of the Wisconsin Supreme Court suggested that judicial elections enhance judicial independence because judicial campaigns provide an opportunity for judges to educate the public about the importance of an independent judiciary that is not present in systems where judges do not have to seek a public mandate. (92) Although no research has been conducted on this specific question, it could be that reappointment by the governor bolsters courts' independence in a manner similar to elections. Because judges who must be reappointed can always claim that they hold their offices with the endorsement of the state's highest elected official, judges that do not face reselection cannot claim to be directly accountable to the people or their elected leaders.

If the suggestion that reselection encourages judges to make more unpopular decisions is correct, it could also be that legislators and governors in states where judges are subject to reselection are more restrained in punishing the judiciary by restricting its powers or funds, or threatening such restrictions. In retention election states, this restraint could be because elections provide judges with an independent source of legitimacy and political clout that judges in states without retention elections do not enjoy. On the other hand, in reappointment states, the restraint could come from the fact that the governor and/or the state legislature do not feel the need to punish the courts with jurisdictional restrictions or budget cuts because they can always remove judges whose decisions they dislike, thereby changing the court's composition.

The results in Table Three suggest the possibly independence-enhancing effect of reselection. The results show that reselection might provide a "counter-weight" to the use of merit selection commissions in the selection process. Reselection might act as a counter-weight by emboldening judges to make more unpopular decisions, despite the fact that they might have been chosen in part because of the belief that they would be biased in favor of the state in criminal cases.

Another interesting finding is that the type of reselection that a state uses for its high court does not appear to affect how often that court ruled in favor of the state. This suggests that justices who must run in retention elections are no more afraid of losing their jobs than justices who must be reappointed. This finding is consistent with the evidence regarding the danger that judges can find themselves in for "pro-criminal" decisions in both systems. Also, the fact that there is no statistically significant difference in the proportion of cases in which the state won that were decided by courts with ten-year terms and those with terms of less than ten years is also somewhat surprising. This suggests that there might not be a relationship between how long justices serve and how a court rules in criminal cases, as long as the justices are subject to some form of reselection. This finding does not refute the conclusions of previous studies (93) indicating that individual judges tend to favor the state more as their terms draw to an end. Given that the justices on state supreme courts serve staggered terms, it would appear that the absolute length of the terms that they serve do not affect the outcome of the decisions they render together.

VI. CONCLUSION

This study reports surprising findings about the tendency of state supreme courts to support the state in criminal cases when comparing the different features of the appointment systems used to select the judges. Above all, this study's results suggest that merit selection might not achieve one of its most important goals: an independent judiciary. Furthermore, the results suggest that subjecting judges to reselection might actually increase their independence, rather than reduce it. In addition, the type of reselection that a state employs, as well as the length of the judicial term, might not be as important to judicial independence as once thought. The results further suggest that forces outside of the normal appointment and reselection process might influence courts' decision-making, specifically in the form of implied or explicit threats from the legislative and executive branches of government to the authority of the judiciary. Judges might take such threats more seriously, or they might receive such threats more often, in states without reselection.

These results also point to the urgent need for more research into how different features of appointment systems affect judicial decisions. Future research should employ analysis techniques to isolate the effects that these features might have on how courts decide cases by controlling for other factors that might affect the outcome of a case (such factors include the overall ideological complexion of the court, the facts and the law at issue in the case, and the crime rate in the state). Such studies, involving more courts and a wider variety of cases, will provide a better and more nuanced understanding of how different appointment system features impact court decision-making.

Studies of the judicial selection processes in several states might shed light on how merit selection commissions vet and select judicial candidates, and how reselection might empower judges to be more independent and have less fear of making unpopular decisions. This research is vital to understanding how the judicial selection process affects the independence of the courts, whose strength and impartiality have important implications for the maintenance of the rule of law.

(1.) AM. JUDICATURE SOC'Y, JUDICIAL SELECTION IN THE STATES 7-14 (2004), available at http://www.ajs.org/js/JudicialSelectionCharts.pdf [hereinafter AM. JUDICATURE SOC'Y, JUDICIAL SELECTION].

(2.) Although all states use either gubernatorial appointments or legislative elections to fill interim vacancies on their highest courts, this study is only concerned with states that use gubernatorial appointment to fill both initial and interim vacancies. See Am. Judicature Soc'y, Judicial Selection in the States, http://www.ajs.org/js/(last visited Oct. 30, 2006) [hereinafter Am. Judicature Soc'y, Judicial Selection in the States].

(3.) Id. The Article uses the term "supreme court" hereinafter to refer to a state's highest appellate court, despite the fact that in some states, the highest appellate court is not officially called the state supreme court. For example, the official name of the highest appellate court in New York is the New York Court of Appeals, while the intermediate appellate court in that state is known as the New York Supreme Court, Appellate Division.

(4.) See F. Andrew Hanssen, Learning About Judicial Independence: Institutional Change in the State Courts, 33 J. LEGAL STUD. 431, 443 (2004).

(5.) Id.

(6.) Id.

(7.) See Kermit L. Hall, The Judiciary on Trial: State Constitutional Reform and the Rise of an Elected Judiciary 1846-1860, 45 HISTORIAN 337, 347, 349-50 (1983) [hereinafter Kermit L. Hall, The Judiciary on Trial].

(8.) Id. at 346-47.

(9.) Id. at 341-54.

(10.) Paul D. Carrington, Judicial Independence and Democratic Accountability in Highest State Courts, 61 LAW & CONTEMP. PROBS. 79, 91 (1998).

(11.) Id. at 91-92.

(12.) Id. at 93; Peter D. Webster, Selection and Retention of Judges: Is There One "Best" Method?, 23 FLA. ST. U. L. REV. 1, 29 (1995).

(13.) Webster, supra note 12, at 29.

(14.) Carrington, supra note 10, at 96.

(15.) Webster, supra note 12, at 29.

(16.) LARRY C. BERKSON & RACHEL CAUFIELD, JUDICIAL SELECTION IN THE UNITED STATES: A SPECIAL REPORT 4-5 (2005), available at http://www.ajs.org/js/ berkson_2005.pdf.

(17.) AM. JUDICATURE SOC'Y, JUDICIAL SELECTION, supra note 1, at 6.

(18.) See AM. JUDICATURE SOC'Y, MERIT SELECTION: THE BEST WAY TO CHOOSE JUDGES 1-3, available at http://www.ajs.org/js/ms_descrip.pdf (last visited Oct. 15, 2006) [hereinafter AM. JUDICATURE SOC'Y, THE BEST WAY TO CHOOSE JUDGES].

(19.) Id. at 1. The governor generally selects commission members, but in some states legislators, the state bar, or members of the state judiciary choose commission members.

(20.) See AM. JUDICATURE SOC'Y, MERIT SELECTION: CURRENT STATUS 3-6 (2003), available at http://www.ajs.org/js/JudicialMeritCharts.pdf [hereinafter AM. JUDICATURE SOC'Y, CURRENT STATUS].

(21.) Id.

(22.) See AM. JUDICATURE SOC'Y, THE BEST WAY TO CHOOSE JUDGES, supra note 18, at 1.

(23.) Id. at 1-3.

(24.) See Aman McLeod, If At First You Don't Succeed: A Critical Evaluation of Judicial Selection Reform Efforts, 107 W. VA. L. REV. 499, 512-14 (2005).

(25.) See Malia Reddick, Merit Selection: A Review of the Social Scientific Literature, 106 DICK. L. REV. 729, 739-40 (2002).

(26.) See PHILIP L. DUBOIS, FROM BALLOT TO BENCH 4 (1980).

(27.) Id.

(28.) See, e.g., Traciel V. Reid, The Politicization of Judicial Retention Elections: Lessons from the Defeat of Justices Lanphier and White, JUDICATURE, Sept.-Oct. 1999, at 70-71.

(29.) See Am. Judicature Soc'y, Judicial Selection in the States, supra note 2. This number includes Alaska and Hawaii, which entered the union in 1959.

(30.) Id.

(31.) See AM. JUDICATURE SOC'Y, JUDICIAL SELECTION, supra note 1, at 6.

(32.) For a full list of states that use merit selection for initial and interim vacancies, see AM. JUDICATURE SOC'Y, CURRENT STATUS, supra note 20, at 3-6.

(33.) Id. at 8-12.

(34.) Id. at 13-15.

(35.) Id.

(36.) Id. at 16-19. For example, commissions differ as to the required degree of confidentiality in their deliberations. Furthermore, some commissions must consider the ethnic and gender diversity of the judiciary when deciding which candidates to forward to the governor.

(37.) See AM. JUDICATURE SOC'Y, JUDICIAL SELECTION, supra note 1, at 7-14.

(38.) Id.

(39.) Id.

(40.) Id.

(41.) Id. Massachusetts, like other states, however, has a mandatory retirement age for judges. Id. at 10.

(42.) Id. at 8.

(43.) For an excellent summary of the empirical research into differences in the decisional behavior of elected and appointed judges, see Reddick, supra note 25, at 745 nn.103-08.

(44.) See, e.g., Carrington, supra note 10, at 91-92; Kermit L. Hall, The Judiciary on Trial, supra note 7, at 347, 349-50.

(45.) See Charles Gardner Geyh & Emily Field Van Tassel, The Independence of the Judicial Branch in the New Republic, 74 CHI.-KENT L. REV. 31, 31 (1998).

(46.) E.g., Robert S. Thompson, Judicial Retention Elections and Judicial Method: A Retrospective on the California Retention Election of 1986, 61 S. CAL. L. REV. 2007 (1988).

(47.) E.g., Reid, supra note 28, at 70-71.

(48.) See, e.g., Joanna Cohn Weiss, Note, Tough on Crime: How Campaigns for State Judiciary Violate Criminal Defendants' Due Process Rights, 81 N.Y.U.L. REV. 1101, 1103-09 (2006).

(49.) See, e.g., Jan Hoffman, A Prominent Judge Retires, Objecting to the Governor's Litmus Test, N.Y. TIMES, Dec. 14, 1997, at 49; Peter Lewis, Rice Won't Reappoint Judge Who Ruled Girl Enticed Attacker, SEATTLE TIMES, Oct. 10, 1990, at A1.

(50.) See Hoffman, supra note 49, at 49.

(51.) See Anthony Champagne, Television Ads in Judicial Campaigns, 35 IND. L. REV. 669, 677-78 (2002).

(52.) Id. at 678.

(53.) See Melinda Gann Hall, State Supreme Courts in American Democracy: Probing the Myths of Judicial Reform, 95 AM. POL. SCI. REV. 315, 324 (2001) [hereinafter Melinda Gann Hall, State Supreme Courts in American Democracy].

(54.) See Gregory A. Huber & Sanford C. Gordon, Accountability and Coercion: Is Justice Blind When It Runs for Office?, 48 AM. J. POL. SCI. 247, 261-62 (2004).

(55.) Id. at 250.

(56.) For example, Hall reports that between 1980 and 1995 only 1.7 percent of supreme court justices who ran in the 234 retention elections during that period were defeated, and that 2.6 percent received less than fifty-five percent of the vote. See Melinda Gann Hall, State Supreme Courts in American Democracy, supra note 53, at 318-19.

(57.) Melinda Gann Hall, Electoral Politics and Strategic Voting in State Supreme Courts, 54 J. POL. 427, 433-35 (1992) [hereinafter Melinda Gann Hall, Electoral Politics].

(58.) Id. at 438-44.

(59.) See Paul R. Brace & Melinda Gann Hall, The Interplay of Preferences, Case Facts, Context and Rules in the Politics of Judicial Choice, 59 J. POL. 1206, 1223 (1997).

(60.) Id.

(61.) See Huber & Gordon, supra note 54, at 255.

(62.) See Paul Brace, Laura Langer & Melinda Gann Hall, Measuring the Preferences of State Supreme Court Judges, 62 J. POL. 387, 398-409 (2000).

(63.) Brace & Hall, supra note 59, at 1221.

(64.) See, e.g., Melinda Gann Hall, Electoral Politics, supra note 57.

(65.) For example, the Democratic presidential candidate won CT, HI, MA, ME, NJ, and RI in 2000 and 2004, while the Republican presidential candidate won AK, CO, TN, and UT in those years. See Fed. Election Comm'n, 2000 Official General Presidential Election Results, http://www.fec.gov/pubrec/2000presgeresults.htm (last visited Nov. 27, 2006); CNN.com, 2004 Presidential Election Results, http://www.cnn .com/ELECTION/2004/pages/results/ (last visited Nov. 27, 2006).

(66.) For a comprehensive description of the judicial selection system that each of these states uses for its supreme court, see Am. Judicature Soc'y, Judicial Selection in the States, supra note 2.

(67.) The cases were gathered from the Lexis-Nexis[c] online database using word searches for published criminal and habeas corpus cases.

(68.) See J. Thomas Sullivan, Unpublished Opinions and No Citation Rules in the Trial Courts, 47 ARIZ. L. REV. 419, 422 (2005).

(69.) See Susan B. Haire, Stefanie A. Lindquist & Donald R. Songer, Appellate Court Supervision in the Federal Judiciary: A Hierarchical Perspective, 37 LAW & SOC'Y REV. 143, 156 n.9 (2003).

(70.) See AM. JUDICATURE SOC'Y, CURRENT STATUS, supra note 20, at 3-6.

(71.) See AM. JUDICATURE SOC'Y, JUDICIAL SELECTION, supra note 1, at 7-14.

(72.) The level of significance measures the certainty that an estimate represents a hypothesized value, which in this case is the difference between the proportion of cases decided by courts with one appointment system feature and those without the feature. The lower the value, the more certain the difference between the two proportions is not zero. See JAY DEVORE & ROXY PECK, STATISTICS: THE EXPLORATION AND ANALYSIS OF DATA 309-11 (1986). For the purposes of this study, significance levels over .95 do not represent a satisfactory level of certainty that there is a difference between two proportions.

(73.) For a more complete discussion of the limitations of studies that focus on published opinions, see Barry Friedman, Taking Law Seriously, 4 PERSP. POL. 261, 271 (2006).

(74.) See Carrington, supra note 10, at 95-97.

(75.) See Melinda Gann Hall, Electoral Politics, supra note 57, at 438-39; see generally Huber & Gordon, supra note 54.

(76.) All of the states in the study used merit selection commissions to select supreme court justices except New Jersey and Maine. See AM. JUDICATURE SOC'Y, JUDICIAL SELECTION, supra note 1, at 6.

(77.) Among the states in the study, Massachusetts and Rhode Island were the only two that do not force their supreme court justices to face some form of reselection. Id. at 10, 12.

(78.) Supreme court justices in the following states in the study serve ten-year terms: Alaska, Colorado, Hawaii, and Utah. Id. at 7-8, 13. Supreme court justices in Connecticut and Tennessee serve eight-year terms, while those in Maine serve seven-year terms. Id. at 8, 10, 13. Supreme court justices in New Jersey serve an initial seven-year term, but face no reselection if renominated by the governor and reconfirmed by the state senate at end of that term. Id. at 11. Because the justices of the New Jersey Supreme Court do not face reselection at regularly recurring intervals, it was excluded from the group of courts whose decisions were compared to see how courts with different term lengths ruled in criminal cases.

(79.) See AM. JUDICATURE SOC'Y, CURRENT STATUS, supra note 20, at 16-17.

(80.) Id.

(81.) ALLAN ASHMAN & JAMES J. ALFINI, THE KEY TO JUDICIAL MERIT SELECTION: THE NOMINATING PROCESS 75 (1974).

(82.) Id. at 239-48.

(83.) Id.

(84.) See, e.g., JOANNE MARTIN, MERIT SELECTION COMMISSIONS: WHAT Do THEY Do? How EFFECTIVE ARE THEY? 20-21 (1993); RICHARD A. WATSON & RONDAL G. DOWNING, THE POLITICS OF THE BENCH AND THE BAR: JUDICIAL SELECTION UNDER THE MISSOURI NONPARTISAN COURT PLAN 101-11 (1969).

(85.) See, e.g., Erwin Chemerinsky, Decision-Makers: In Defense of Courts, 71 AM. BANKR. L.J. 109, 113-14 (1997).

(86.) See, e.g., J. W. PELTASON, FIFTY-EIGHT LONELY MEN: SOUTHERN FEDERAL JUDGES AND SCHOOL DESEGREGATION 9-12 (1961); Julian Borger, Former Top Judge Says U.S. Risks Edging Near to Dictatorship, THE GUARDIAN, Mar. 13, 2006, available at http://www.guardian.co.uk/usa/story/0,,1729396,00.html (discussing a possible connection between violence against judges and judges' decisions).

(87.) E.g., Gary M. Anderson, William F. Shughart II & Robert D. Tollison, On the Incentives of Judges to Enforce Legislative Wealth Transfers, 32 J.L. & ECON. 215, 226 (1989); Eugenia Froedge Toma, Congressional Influence and the Supreme Court The Budget as a Signaling Device, 20 J. LEGAL STUD. 131, 145-46 (1991).

(88.) See, e.g., Shirley S. Abrahamson, Judicial Independence as a Campaign Platform: The Importance of Fair and Impartial Courts, 84 MICH. B.J. 40, 42 (2005) [hereinafter Abrahamson, Judicial Independence].

(89.) See, e.g, Roy B. Flemming & B. Dan Wood, The Public and the Supreme Court: Individual Justice Responsiveness to American Policy Moods, 41 AM. J. POL. SCI. 468, 493-94 (1997).

(90.) See generally Kermit L. Hall, The Judiciary on Trial, supra note 7.

(91.) See Larry V. Starcher, Choosing West Virginia Judges, 20 QUINMPIAC L. REV. 767, 769-70 (2001).

(92.) See Shirley S. Abrahamson, The Ballot and the Bench, 76 N.Y.U. L. REV. 973, 990 (2001); Abrahamson, Judicial Independence, supra note 88, at 41-42.

(93.) See generally Melinda Gann Hall, Electoral Politics, supra note 57; Huber & Gordon, supra note 54.

Aman L. McLeod, Assistant Professor of Political Science and Law, Rutgers University--Camden; Ph.D., J.D., University of Michigan; B.A., Amherst College. The author would like to thank Carmen Adamucci, Katie Blasi, Victoria Haich, Jacob Rosoff, Danielle Reiss, Melissa Ruggeri, and Jason Toy for their help gathering the data for this study. The author would also like to thank Norman Greene, Russell Harrison, Bettie McLeod, Peter Palenchar, and G. Alan Tarr for their helpful comments.

TABLE ONE
NUMBER OF CRIMINAL CASES AND OUTCOMES
FOR STATE SUPREME COURTS IN THE SAMPLE

                           State Government   State Government
State                           Loser              Winner        Totals

Alaska
No. of Cases                      11                 19            30
% of Alaska Cases              (36.7%)            (63.3%)        (100%)

Colorado
No. of Cases                      48                 90           138
% of Colorado Cases            (34.8%)            (65.2%)        (100%)

Connecticut
No. of Cases                      34                133           167
of Connecticut Cases           (20.4%)            (79.6%)        (100%)

Hawaii
No. of Cases                      33                 36            69
% of Hawaii Cases              (47.8%)            (52.2%)        (100%)

Maine
No. of Cases                      40                103           143
% of Maine Cases               (28.0%)            (72.0%)        (100%)

Massachusetts
No. of Cases                      90                314           404
% of Massachusetts Cases       (22.3%)            (77.7%)        (100%)

New Jersey
No. of Cases                      54                 66           120
% of New Jersey Cases           (45%)              (55%)         (100%)

Rhode Island
No. of Cases                      25                187           212
% of Rhode Island Cases        (11.8%)            (88.2%)        (100%)

Tennessee
No. of Cases                      43                 73           116
% of Tennessee Cases           (37.1%)            (62.9%)        (100%)

Utah
No. of Cases                      13                 76            89
% of Utah Cases                (14.6%)            (85.4%)        (100%)

Totals
No. of Cases                     391                1097          1488
% of Total Cases               (26.3%)            (73.7%)        (100%)

TABLE TWO
DIFFERENCES IN STATE SUPREME COURT RULINGS WITH
DIFFERENT APPOINTMENT SYSTEM FEATURES

                                             Standard
Appointment System Feature    N     Means   Deviations

Merit Commission             1225    .76       .429
No Merit Commission           263    .64       .480

Reselection                   872    .68       .465
No Reselection                616    .81       .390

Ten-Year Term                 326    .68       .468
Less than Ten-Year Term       426    .73       .447

                             Difference of
Appointment System Feature   Means Test *    Significance

Merit Commission
No Merit Commission              3.588           .000

Reselection
No Reselection                   5.834           .000

Ten-Year Term
Less than Ten-Year Term         -1.405           .161

* Test does not assume equal variances.

TABLE THREE
EFFECTS OF MERIT SELECTION COMMISSIONS ON
COURTS' DECISIONS

                                            Standard
Appointment System Feature    N    Means   Deviations

Merit Commission Without
Reselection                  616    .81       .390

Merit Commission With
Reselection                  609    .70       .458

Merit Commission With
Reselection                  609    .70       .458

Reselection Without Merit
Commission                   263    .64       .480

                             Difference of
Appointment System Feature   Means Test *    Significance

Merit Commission Without
Reselection                      4.612           .000

Merit Commission With
Reselection

Merit Commission With
Reselection                      1.676           .094

Reselection Without Merit
Commission

* Test does not assume equal variances.

TABLE FOUR
THE EFFECTS OF RESELECTION ON COURTS' DECISIONS

                                              Standard
Appointment System Feature      N    Means   Deviations

Retention Election             373    .69       .462
No Reselection                 616
Retention Election             373    .69       .462
Reselection by Reappointment   499
Reselection by Reappointment   499    .68       .468
No Reselection                 616

                               Difference of
Appointment System Feature     Means Test *    Significance

Retention Election                -4.247           .000
No Reselection
Retention Election                  .451           .652
Reselection by Reappointment
Reselection by Reappointment       5.192           .000
No Reselection
* Test does not assume equal variances.