State Supreme Court Decision Making in Confession Cases
Benesh, Sara C., Martinek, Wendy L., Justice System Journal
State Supreme Court Decision Making in Confession Cases*
The federal nature of the American judiciary suggests that a state court of last resort may evade decisions of the U.S. Supreme Court if those decisions do not comport with the preferences of the state supreme court judges or are in conflict with the prevailing ethos in the state. We offer a multiple principal agency model of state supreme court decision making. We posit that the decisions of state high courts are influenced by their judicial (the U.S. Supreme Court) and political (state elites or electorates) principals, as well as by more conventional factors. We test our theory by using a stratified random sample of state court of last resort decisions regarding challenged confessions from 1970 to 1991. Our analysis supports the hypothesized influence of federal courts on state supreme courts. That influence transcends most of the known determinants of decision making on the state supreme courts. We conclude that state supreme courts defer to their judicial principal but do not hesitate to use federalism to their advantage. In this area of the law, though, they do so without compromising Supreme Court precedent.
The federal nature of the American system of government is not the product of coincidence or chance but, rather, is rooted in the historical experiences of the United States. Suspicion of centralized power, a hallmark of colonial experience, gave rise to a system of government with a special concern for state sovereignty. One result of particular interest to students of the courts is the existence of individual state court systems, each operating independently but tied to the federal judiciary via the supremacy clause of the Constitution. When a question related solely to the interpretation and application of a state constitution or state statutory provision arises, according to the precepts of federalism, the federal judiciary is virtually silent. In these cases, the state courts of last resort reign supreme.1
When the issue is a matter of federal law or the application of the federal Constitution, however, the United States Supreme Court is formally the final arbiter, and state supreme courts are considered bound by the applicable rulings of the nation's highest tribunal. In fact, the Supreme Court, speaking through Justice Sandra Day O'Connor, made a strong statement to this effect in its decision in Michigan v. Long:
It is precisely because of this respect for state courts, and this desire to avoid advisory opinions, that we do not wish to continue to decide issues of state law that go beyond the opinion that we review, or to require state courts to reconsider cases to clarify the grounds of their decisions. Accordingly, when, as in this case, a state court decision fairly appears to rest primarily on federal law, or to be interwoven with the federal law, and when the adequacy and independence of any possible state law ground is not clear from the face of the opinion, we will accept as the most reasonable explanation that the state court decided the case the way it did because it believed that federal law required it to do so.2
While this opinion may be interpreted as favoring a strong central government, it also provides state supreme courts with the opportunity to avoid review by resting their analysis on independent state grounds. We might expect, then, that state courts of last resort will take advantage of this invitation to use federalism to their own advantage, allowing them to decide cases incompatibly with the United States Supreme Court.
With this in mind, we attempt to further understand the decision making of state supreme courts. Because these courts are not only the heads of their own state court systems but also ostensibly members of the federal judicial system, we consider the influence of the United States Supreme Court in the explanation of state supreme court behavior. We draw on principal agency theory to explain the relationship between superiors (the United States Supreme Court) and inferiors (state supreme courts) in an empirical study of a sample of confession cases decided by state supreme courts between 1970 and 1991. We extend the work of those scholars who have argued that understanding state supreme courts requires placing them in their state contexts (e.g., Hall and Brace 1999) by arguing further that understanding state supreme courts also requires cognizance of their relationship to the United States Supreme Court.
Principal Agency Theory
Some scholars have argued that the Supreme Court as an institution really has very little impact and cannot single-handedly change the direction of any given policy (Rosenberg, 1991). However, recent empirical work has demonstrated that both federal courts (Songer, Segal, and Cameron, 1994; Benesh, 2002) and state supreme courts (Martinek, 2000; Songer, 2000) do pay at least some heed to the Supreme Court. This is of interest if for no other reason than the fact that the Supreme Court is an impotent organization indeed if even the courts within its own hierarchy-directly, with the federal courts, or indirectly, with the state supreme courts-do not concern themselves with its decisions. The formally hierarchical relationship of the United States Supreme Court to state courts of last resort suggests that principal agency theory can shed some light on the relationship between the two.
Principal agency theory was first developed in economics as a means of understanding the relationship between buyers and sellers (e.g., Ross, 1973), each presumed to have different goals. Principals (buyers) wish to receive the highest quality services from agents (sellers) at the lowest possible price, while agents wish to maximize profit, which could include selling more services than the principal needs or maximizing price while minimizing quality. Principal agency theory made its way into political science as a tool for understanding bureaucratic performance (e.g., Mitnick, 1975, 1980), with legislators as the principals and bureaucrats as the agents. Principal agency models, with appropriate modifications, are also well suited to the study of the relationship between superior and inferior courts.
In the classic formulation of principal agency theory as applied to the bureaucracy, bureaucratic agents are able to shirk-that is, direct their efforts toward nonpolicy goals (Brehm and Gates, 1997:21-because of a fundamental information asymmetry in which the agents have information the principal does not possess. In short, bureaucratic agents know more about their own skills and wants and about the tasks they are charged with accomplishing than do the principals who assign those tasks. Principals must devote considerable resources to monitoring the behavior of their agents to ameliorate this asymmetry, but because of the cost of monitoring, the information asymmetry is rarely if ever completely eliminated.
Two key concepts from principal-agent theory are relevant to information asymmetry. One is the notion of "adverse selection," in which the principal cannot know the agents' true preferences and abilities. The other is the notion of "moral hazard," which means that the principal cannot observe all actions of subordinates (see Brehm and Gates, 1997:19). With respect to adverse selection, the United States Supreme Court faces a unique situation with regard to state courts of last resort. Even if the Supreme Court possessed perfect information as to state supreme court judges' preferences, it would be unable to use this information to staff state supreme court benches with jurists likely to be faithful because the Court has nothing to do with their selection. As to moral hazard, the number of cases state supreme courts decide, relative to the number disposed of by the United States Supreme Court, is staggering. Thus, the Supreme Court faces a substantial challenge in monitoring the output of state courts of last resort. For example, from 1983 to 1994, the yearly average number of cases disposed of formally by a state supreme court was 563 (National Center for State Courts, 1995), which, when multiplied by the fifty-two state courts of last resort, totals some 29,276 decisions. This leaves much the United States Supreme Court simply cannot monitor. Adding to this difficulty is the expanding number of cases decided by the United States Courts of Appeals-54,963 cases in 1999 (Administrative Office of the Courts, 2000)-which are also important Supreme Court agents.
Indeed, Brisbin and Kilwein (1994) report that relatively few state supreme court cases are reviewed in any Supreme Court term. Further, of those cases reviewed by the Court, typically no more than a third are reversed outright. The fact that the United States Supreme Court cannot choose to hear cases that are not appealed to it exacerbates the moral hazard problem. Even if the Court had time to monitor state supreme court decision making effectively, it would still depend on other actors to draw its attention to cases it ought to review.
The rich literature on judicial politics and behavior suggests several other constraints to which heed must be paid. Those emphasizing the role of attitudes as an explanation for judicial behavior (the attitudinalists--Segal and Spaeth, 1993) warn we must include ideology, while others (known as the new institutionalists-see Hall and Brace, 1999) direct our attention to institutional structures, especially their effect on the expression of attitudes. Collectively, this work suggests the need to allow for the presence of multiple principals when we consider state courts. We offer a model of multiple principals that includes legal (Supreme Court precedent and the constitutions of the states), attitudinal (the policy preferences of the state supreme court jurists), and institutional (docket control and workload considerations) elements. We also pay heed to the relevant actors in the states (the electorate or the state elites) and, thus, to state political environments that may serve to mediate the relationships among the factors just noted. We look beyond the context of highly salient issues (such as the death penalty-Hall and Brace, 1992, 1994; Emmert and Traut, 1994; Brace and Hall, 1995; Traut and Emmert, 1998), examining various explanations for judicial decisions in confession cases.
State Supreme Courts in Context
The role of context in judges' decisional calculus applies not only to the United States Supreme Court (e.g., Epstein and Walker, 1995; Segal, 1997), but also-and probably more aptly-to state courts of last resort (e.g., Emmert and Traut, 1994; Brace and Hall, 1995). While most scholars agree with the canon that attitudes play some role in judges' decision making, scholars also recognize that "any attempt to explain behavior with reference to beliefs but not context such as institutional settings will inevitably be incomplete:' as contextual factors make enacting policy preferences either more or less difficult and may even create certain goals and preferences related to the institution itself (Gillman and Clayton, 1999:3-5). Multiple influences especially apply to the situation in which judges of state courts of last resort find themselves. State supreme courts do not operate in a vacuum. Rather, they are part of a political landscape populated by other institutions, citizens, and elites. This suggests that state supreme courts have political principals (state electorates, state political elites) in addition to their judicial principal (the United States Supreme Court).
The notion of multiple principals gives rise to a multifaceted role for ideology. To begin with, because of the broad consensus in the field that ideology matters not only on the Supreme Court (e.g., Rohde and Spaeth, 1976; Segal and Spaeth, 1993), but also on the Courts of Appeals (e.g., Songer and Haire, 1992; Songer, Segal, and Cameron, 1994; Reddick, 1997) and on the state supreme courts (e.g., Emmert and Traut, 1994, Flemming, Holian, and Mezey, 1998), state supreme court judges presumably attempt to enact their most preferred policies. Consequently, in the context of confession cases, ceteris paribus, liberal courts should vote more often to exclude confessions than conservative courts. However, the role of ideology is not likely to end there.
The Supreme Court's ideology sends signals to lower courts about the probability of review and reversal. The ideology of the Supreme Court, then, becomes a proxy for the likelihood of reversal, and as we generally argue that judges do not like to be reversed (Baum, 1978:212-13), this probability of reversal should influence decision making. Assuming lower courts do not like to be overruled, a lower court should look first to the United States Supreme Court to see whether its decision is likely to be overturned. As the Supreme Court becomes more conservative, the likelihood of the reversal of liberal decisions (e.g., invalidating challenged confessions) increases. Obviously, the same relationships hold, in the opposite direction, for conservative decisions.
In addition, state supreme courts have more than just their judicial principal (the United States Supreme Court) to consider. They must also be cognizant of their political principals: those actors responsible for their retention on the bench. State supreme court jurists ascend to their positions through a variety of appointive and elective mechanisms. In either case, dominant political values in the state come into play. Scholars have demonstrated that elected judges display a definite sensitivity to their constituencies' preferences (e.g., Hall, 1992; Emmert and Traut, 1994; Hall and Brace, 1996; Traut and Emmert, 1998). Just as elected judges are likely to act with the electorate's preferences in mind, if they are strategic (which the extant evidence suggests they are--see Hall, 1992; Brace, Hall, and Larger 1996; Larger, 1997), there is reason to believe that when appointed judges act, they are likely to keep in mind the preferences of the political elites responsible for their accession to, and retention on, the bench. Indeed, in a recent analysis, Larger (1999) found that members of state supreme court benches do show concern for retaliation by other state political actors (the legislature and governor), as manifested by a lessened propensity to engage in judicial review, for at least some areas of law. We argue, relatedly, that justices will take ideology into account when deciding confession cases, making decisions to overturn confessions only when the ideology of the citizenry or the elites (depending on selection system) is liberal enough to support such a decision.
The new institutionalists have provided persuasive evidence in support of the thesis that institutional arrangements matter for judicial outputs. For example, some state supreme courts are relieved of the burden of a nondiscretionary docket, or at least have that burden ameliorated; others, however, must hear every case that comes to them, which lessens their ability to effectuate policy. We presume that the existence of an intermediate appellate court and a discretionary (rather than mandatory) criminal jurisdiction of a state supreme court indicates at least some freedom from frivolous cases. When state supreme courts can rely upon intermediate appellate courts to resolve "easy" cases, they can restrict themselves to the more important ones, providing them with the opportunity to decide more cases in which there may be a constitutionally defective confession. Having an intermediate appellate court also reduces the workload for the state supreme court, giving judges additional room to make policy and take "big" cases. So, too, discretion with regard to criminal appeals allows judges to focus on policy content in selecting the cases they will hear. Therefore, we expect that, ceteris paribus, state supreme courts with an intermediate appellate court and those with discretionary criminal jurisdictions will be more likely to overturn challenged confessions.
A state's constitution is likely to be relevant, as well. As one state supreme court jurist remarked, "[S]tate charters offer important local protection against the ebbs and flows of federal constitutional interpretation:13 In most instances states have adopted, more or less word-for-word, the language of the United States Constitution's Bill of Rights in guaranteeing protection against self-incrimination (the protection relevant to confession cases). There are, however, a few instances in which the state constitution is more protective of individual rights than the federal Constitution and others where it is less so. Presumably, the language of the state constitution should affect the decisions a state supreme court jurist can make. When the state constitution provides more expansive protections than those afforded by the federal Constitution, it enhances the likelihood that a confession will be invalidated as involuntary. Conversely, when the state constitution provides less expansive protections than those afforded by the federal Constitution, the likelihood of invalidating a confession as involuntary is likely to be diminished.
State courts also have the liberty to rely upon independent state grounds in writing their opinions. We expect that state courts may plausibly avoid the thrust of United States Supreme Court doctrine by resting decisions on independent state grounds, as Michigan v. Long made clear they could do. Accordingly, those state courts operating under more protective state constitutions are expected to be more likely to exclude a confession when reliance is placed on state rather than federal grounds than those with less protective state constitutions.
Considered collectively, these variations in external environment, institutional features, and legal context allow for a more precise test of the extent to which state supreme courts are truly faithful to their judicial and political superiors. By examining a host of legal, attitudinal, and institutional factors, we gain insight into the faithfulness of these courts to their multiple principals.
The Confession Cases
To advance our understanding of state supreme court decision making in particular, and judicial decision making more generally, we began with a systematic analysis of the jurisprudence of the United States Supreme Court in the area of voluntary confession to ascertain whether state supreme courts relied on the same factors noted in Supreme Court cases to dispose of their confession cases. If many of the factors identified by the Supreme Court also affect the likelihood that a state court will either accept or reject a confession, then we have at least preliminary evidence that state supreme courts pay attention to the Supreme Court's decision making.
Confession cases fit the analysis employed here for several reasons. In this area of law, differences in fact contribute to the Supreme Court's decisions, as the justices themselves have indicated (Haynes v. Washington"). This area of law thus allows us to determine whether the lower courts and the Supreme Court consider the same facts to be important. These facts, which proceed from Supreme Court precedent, embrace defendant characteristics, as well as tactics employed by law enforcement to obtain confessions. In analyzing Supreme Court decisions and their relevant facts, we draw extensively from Benesh's (2002) application of a fact model of confession cases to the decisions of the United States Courts of Appeals. She demonstrated the attention those courts paid to the Supreme Court. Relying on Kort's (1963) analysis, as well as on several sources on confessions and their admissibility (e.g., Inbau and Reid, 1962; Schafer, 1968; Rutledge, 1994; Grano, 1996) and the Supreme Court's decisions themselves, she derived the relevant facts pertinent to the decisions on the admissibility of confessions, which we posit should also influence decision making in state supreme courts.
First, actual coercion is an important factor in the determination of voluntariness. According to the Supreme Court's rulings in Brown v. Mississippi and Lee v. Mississippi, where coercion (either physical or psychological) is present, a court should be more likely to overturn confessions. Coercion other than explicitly physical or psychological coercion-including deprivation of basic needs, lengthy detentions, and holding of a suspect incommunicado-is also likely to enhance the likelihood of a confession being overturned (see, e.g., Davis v. North Carolina and Payne v. Arkansas). Further, if a confession is preceded by another, involuntary confession, the likelihood of exclusion should increase (Silverthorne Lumber Co. v. U.S.).
Second, a number of characteristics of the accused may lead that individual to be more or less influenced by coercive methods, and these characteristics are, thus, likely to affect a court's treatment of a challenged confession. These include mental status, intelligence, race, experience with the legal system, and youth (see, e.g., Chambers v. Florida; Culombe v. Connecticut, Gallegos v. Colorado; Spano v. New York). Other miscellaneous characteristics that are also potentially influential include being addicted to drugs, being under the influence of alcohol, and being ill.
Another set of factors pertains to procedural defects that might taint the voluntariness of a confession and, thus, enhance the prospects a court will exclude it. Examples include the failure to warn a defendant of Miranda rights, lengthy interrogations (Turner v. Pennsylvania), use of police relay tactics during questioning (Lisenba v. California), and use of the fruits of illegality, such as information stemming from an illegal. arrest or search (Wong Sun v. U.S.). Other procedural elements include courtroom procedural fairness, including whether the trial court heard testimony outside the presence of the jury as to voluntariness (Jackson v Denno), the effect of a codefendant's confession on the defendant's confession (Bruton v. U.S.), and failure to bring a defendant before a magistrate in a timely fashion (McNabb v. US. and Mallory v. US.). There is also a set of pertinent mitigating circumstances-those likely to enhance the prospects of a challenged confession being upheld-to be considered. These include a knowing and intelligent waiver of rights (Miranda v. Arizona) and application of the harmless error doctrine (Milton v. Wainwright).
Each of these confession case facts represents the influence of United States Supreme Court precedent. Most of the other Supreme Court confession cases during our period of study (1970-1991) involved certain combinations of these facts, which led either to the admission or to the exclusion of a confession and, therefore, did not present the lower court with any broad new precedent to apply. However, the Supreme Court's reaffirmation of certain facts may well lead state supreme courts to pay greater heed to the absence or presence of these facts. Because the Supreme Court can fully choose its agenda, we posit that repeated decisions in cases involving specific facts can be seen as prioritizing behavior. Thus, we submit that the Supreme Court cares most about the presence or absence of those facts it references most often and that the lower courts are cognizant of that message. To take this into account in the analysis that follows, we consider only those confession case facts that were present in at least one-third of Supreme Court decisions: Miranda warnings given in whole or in part, psychological coercion, length of interrogation, incommunicado detention, and magistrate hearing.5 (See Table 1 for the distribution of facts in Supreme Court decisions.) The number of case facts present in any case that support admission of a confession (e.g., Miranda rights were given) as compared with the number of case facts present that tilt toward excluding the confession (e.g., the suspect was subjected to psychological coercion) will be important. Therefore, tracking only the presence of the Supreme Court's most cited factual considerations, we measure the influence of precedent as the difference between positive (those that support admission of the confession) and negative (those that support exclusion of the confession) case facts.6
This measure of precedent allows us to evaluate the compliance of state supreme courts even though Supreme Court precedent is relatively stable during our period of study. While most studies of compliance focus on highly salient areas of law characterized by volatility in Supreme Court precedent, we are able to leverage a greater understanding of compliance more generally by purposefully selecting a less salient and more stable area of law. This necessitates a measure of precedent that is not dependent upon changes in the Supreme Court's jurisprudence, which is exactly what our measure is.
The foregoing discussion has identified a host of factors likely to be at play in state supreme court decision making. Our hypotheses fall into several categories. First is the basic attitudinal hypothesis. Second is the hypothesis dealing with the influence of the United States Supreme Court, the judicial principal for state supreme courts. Third is the hypothesis dealing with the influence of those courts' political principals: the state electorate or state political elites. Finally are hypotheses regarding institutional considerations that may impinge on state supreme court decision making.
The Attitudinal Hypothesis
H1: The more liberal a state supreme court, the more likely it is to exclude a challenged confession.
The Judicial Principal Hypothesis
H2: The more liberal the United States Supreme Court, the more likely a state supreme court is to exclude a challenged confession.
The Political Principal Hypothesis
H3: The more liberal the state context (the electorate in elective states, political elites in appointive states), the more likely a state supreme court is to exclude a challenged confession.
The Institutional Hypotheses
H4: In a state with an intermediate appellate court, the state supreme court is more likely to exclude a challenged confession.
H5: A state supreme court with discretionary jurisdiction in criminal cases is more likely to exclude a challenged confession.
H6: A state supreme court operating under a state constitution that is more protective of criminal defendants than the United States Constitution is more likely to exclude a challenged confession.
H7: A state supreme court operating under a state constitution that is more protective of criminal defendants than the United States Constitution and that rests its opinion on independent state grounds is more likely to exclude a challenged confession.
Our desire to assess the complex web of relationships with which state supreme courts must contend in rendering their decisions led us to select an area of case law that, while important, differs from the highly salient cases that have served as the focus of most studies of state supreme court decision making. While there are, no doubt, several to choose from, as has already been noted, we focused on cases in which the voluntariness of a confession is challenged. The sample of confession cases for this study was constructed by identifying all 4,008 cases cited by West's Key Number System (Keys 516538) and the Decennial Digest for the years 1970-1991 as involving a confession. From this universe, a random sample of 800 cases was selected, stratified by state (see Table A-1, Appendix A). Each of these cases was then read in its entirety. We excluded those cases not involving the voluntariness of a confession and those where the confession was merely discussed but the appellant did not question its voluntary nature.7 A final sample of 661 confessions resulted.8 A few cases involved challenges to more than one confession (either two or more separate confessions by the same individual or separate confessions by two or more individuals), thus yielding more than one observation.
We chose the 1970-1991 period for several reasons. First, because our objective was to understand state supreme court decision making across the fifty states, we needed to select a reasonable length of time that allowed us to include a representative number of cases from each state and year. Second, essential to our conception of state supreme court decision making is the role of ideology, both of the state supreme court itself as well as of the political and judicial principals involved. In the case of state supreme court preferences and those of state citizenries and elites, the best measures of those preferences are those of Brace et al. (2000) and Berry et al. (1998), respectively. These measures were available for the 1960-1993 period and, hence, were available for our selected period of study. Finally, the Supreme Court's jurisprudence during this period was relatively stable. Most studies of compliance have focused on periods characterized by instability and changeability in Supreme Court precedent. Incidents of compliance (or noncompliance, as the case may be) during such periods are undoubtedly important tests of the influence of the United States Supreme Court. Our interest, however, is in contributing toward an understanding of compliance more generally. Just as we have chosen to focus on an important but less highly charged issue (confession cases rather than abortion or death penalty cases), we have also chosen to focus on compliance during a period of relative stability in the law.
Operationalization of Variables
In coding the data, we separated each confession case fact into two elements: 1) whether or not the fact was mentioned as present and 2) whether or not the fact was mentioned as not being present.9 This provides additional leverage to evaluate the Supreme Court's importance as the judicial principal because attention to its decision making is demonstrated in either case. If the fact is mentioned as not being present, that very mention implies that the presence of it matters, even though the factor is not present in the case at hand. Because there are so many case facts and because so many state supreme court decisions mention no confession-related fact at all (see Table 2), we focus on the facts cited most frequently by the United States Supreme Court. Recall that our measure for precedent is the difference between the number of Supreme Court frequently cited facts that would lead to the admission of a confession and those that would lead to its exclusion. Hence, positive values should be associated with a lesser likelihood of invalidating a challenged confession, while negative values should be associated with a greater likelihood of invalidating a challenged confession. It is duly noted that this way of measuring Supreme Court precedent is imperfect and probably dilutes the impact of individual Supreme Court decisions. Still, overall, cases that include facts often mentioned by the Supreme Court should be treated differently, if the lower courts are complying with the high court, and so the difference between mentions of positive and negative facts is pragmatic and meaningful (see also note 6).
Ideology of the state supreme court is calculated as the mean party-adjusted ideology (PAJID) score (Brace, Larger, and Hall, 2000) of the state supreme court judges in the majority in each case. United States Supreme Court ideology is the mean Segal/Cover score for the sitting Supreme Court in the year in which the decision is rendered (Segal and Cover, 1989; Segal et al., 1995). We adapted Songer's (2000) novel approach to measure the preferences of political principals. Songer used the Berry et al. (1998) elite ideology measure as a measure of the dominant political values at the time a state supreme court judge came to office in cases in which judges were appointed. He used the Berry et al. (1998) citizen ideology as the measure of the dominant political values at the time a state supreme court judge came to office in cases in which judges were elected. Songer's interest was in tapping into judicial values, using the elite and citizen ideology variables as surrogates. We are directly interested in the political preferences of the relevant state actors: elites when retention in office is dependent upon the governor, legislature, or both, citizens when retention is dependent upon the electorate. Hence, we measure the preferences of political principals as the Berry et al. (1998) elite ideology score at the time a decision is rendered in states in which retention in office is dependent upon reappointment by the governor, legislature, or both. In those states in which retention in office is via an electoral mechanism (merit retention, partisan election, nonpartisan election), the measure is the Berry et al. (1998) citizen ideology score at the time a decision is rendered. We ascertained the method of retention by reference to the Book of the States (Council of State Governments, various years). For each ideology measure, higher values represent more liberal ideological positions while lower values represent more conservative ones.
To ascertain the level of self-incrimination protection afforded by state constitutions, the text of each state constitution was searched via Lexis-Nexis and compared to the federal exemplar to gauge the similarity of their relevant provisions. 10 Independent state grounds were coded on the basis of the text of the state supreme court decision. Most relevant with regard to the grounds cited by the state supreme court is the interaction of state grounds with state constitutional provisions. Relying solely on independent state grounds in the context of greater state constitutional protections can provide greater latitude for a state supreme court to render a more liberal outcome (and vice versa).
Information on the presence or absence of an intermediate appellate court, gathered from the Book of the States (Council of State Governments, various years), was supplemented by consulting the Web sites of state courts of last resort and making phone calls to state court administrative offices. Information on the discretionary versus mandatory nature of state supreme court criminal jurisdiction was obtained from the Bureau of Justice Statistics (1998).
We also included two variables essentially as controls. First, we indicated whether the underlying criminal offense was murder. Arguably, judges are less likely to side with criminal defendants when more serious offenses are involved, and murder is arguably the most serious offense. Second, we indicated whether the lower court upheld or invalidated a challenged confession. This controls for the overwhelming propensity of the state supreme courts to affirm the decisions of the trial and intermediate appellate courts.
Analysis and Results
The dichotomous nature of our dependent variable (whether or not a challenged confession was excluded by the state supreme court) makes logit an appropriate estimation technique (see Table 3).11 Perhaps the most interesting, and surprising, finding is that, when controls for a host of other relevant factors are included, the preferences of the state supreme court decision-making majority do not matter much, failing to achieve statistical significance at the conventional 0.05 level. This result differs from other studies of state supreme courts (e.g., Martinek, 2000) and is truly remarkable in light of the tenacity of the attitudinal model (which argues that ideology is the predominant determinant of judges' voting). This result may reflect the absence of a strong ideological content in voluntary confession cases, an explanation corroborated by our findings with regard to the influence of the various principals to which state supreme courts are subject. The level of significance (0.058) does suggest some room for ideology influence, however.
We find absolutely no evidence to support the notion that either the ideological preferences of the Supreme Court or the preferences of state political principals matter. This latter finding runs contrary to much of the literature devoted to state supreme court decision making and suggests that the state courts are quite independent in their decision making, at least ideologically. Their own ideology may matter somewhat, but the ideology of those superior to them legally, and of those to whom they owe their tenure, matters not at all. Either neither is a real threat, or neither has control over state supreme court behavior.
While we find no evidence in favor of the influence of Supreme Court ideological preferences, we do find powerful evidence for the influence of Supreme Court precedents. It appears, then, that the state supreme courts are using Supreme Court precedents without reference to the high court's ideology. This suggests that they are not looking to predict reversal or review but, rather, are seeking to decide cases in accordance with the legal guidance contained in Supreme Court voluntary confession cases.
When we turn to institutional effects, we find mixed results. To begin with, it does not matter whether there is an intermediate appellate court. However, having a discretionary criminal docket is associated with an increase in the likelihood of a challenged confession being excluded; a discretionary criminal docket is associated with an 8 percent greater likelihood of invalidating a challenged confession than a mandatory docket. This comports with the notion that such discretion will result in the docketing of cases containing stronger arguments in favor of the defendant-opportunities for policymaking. We also find, as anticipated, that when a state rests its decision on independent state grounds, it is less likely to exclude a confession. While the state supreme courts indeed follow Supreme Court precedent, at least in terms of the facts the Court has identified as affecting the voluntariness of confessions, there are apparently other cases in which the states use their own precedents or constitutional provisions to allow confessions that may be otherwise questionable.
State constitutional provisions also affect the likelihood a confession will be excluded. Here, however, we find an anomaly. We hypothesized that state constitutional provisions more protective of a defendant's right against self-incrimination would contribute to a greater likelihood of excluding a challenged confession. However, we find that the relationship runs in the opposite direction, with more protective state constitutional provisions associated with a decreased likelihood of exclusion. Our admittedly speculative explanation is that a more protective state constitution may invite criminal defendants to appeal adverse decisions, resulting in less than compelling legal arguments for the exclusion of the confession. When we couple a decision made on independent state grounds with more protective state constitutional provisions, however, we do find the result expected. When a state high court rests its decision on independent state grounds, the more protective the state's own constitution, the more likely the court is to exclude a confession.
Both of our control variables have the expected effect. When the underlying case involves a murder charge, the likelihood of the confession being invalidated is depressed-however, it fails to meet conventional standards for significance, so its influence is not certain. But when the lower court has invalidated a challenged confession, the likelihood of the state supreme court also invalidating the confession is enhanced.
To better illustrate the substantive effects of these independent variables, we generated an array of simulated probabilities based on different values of the independent variables (see Table 4). Consider first the influence of Supreme Court precedent on state supreme court decision making. The influence is remarkable. When we find the maximum difference between the number of confession case facts likely to lead to excluding the confession and the number of confession case facts likely to lead to allowing the confession (four more of the former than the latter), the likelihood of exclusion is 87 percent. Compare this to the expected probability when we have the maximum difference between case facts likely to lead to allowing the confession and those likely to lead to excluding the confession (six more of the former than the latter): 2 percent.
The effect of a mandatory versus discretionary criminal jurisdiction is not as stark in terms of the substantive effect as United States Supreme Court precedent, but neither is it trivial. With a mandatory criminal jurisdiction, and all other variables set at their mean or modal values, the likelihood of excluding a confession is 10 percent. When, however, the state court of last resort enjoys a discretionary criminal jurisdiction, that probability goes to 18 percent.
Our results are at odds with several established tenets in the judicial politics scholarship; specifically, neither the attitudinal nor the new institutional theories of judicial decision making travel well to the area of voluntary confession cases in the state supreme courts, while legal factors matter a great deal. With regard to the confession cases, consider first that United States Supreme Court precedent does matter. Indeed, it and the direction of the lower court's decision are the most influential factors for state supreme court decision making in confession cases. So, while the state courts overwhelmingly affirm the cases coming from their trial or intermediate appellate courts, they do so in keeping with Supreme Court precedent on point. This suggests, of course, that the trial courts and intermediate appellate courts themselves are following Supreme Court precedent; perhaps these lower courts are the true agents. The ramification for federalism here is clear: while the state supreme courts may appear to be only formally subordinate to the United States Supreme Court, in practice they do heed their federal judicial principal.
We also find state constitutions to be influential, when coupled with judicial use of independent state grounds for rendering a decision. When the state constitution is less protective and the state court of last resort rests its decision on state grounds, the likelihood of exclusion is diminished; when the state constitution is more protective and the state court of last resort rests its decision on state grounds, the likelihood of exclusion is enhanced. Again, the ramifications for federalism are clear, as we find that the states, when left to their own accord, can either do better than the federal government or do worse. And it matters what they put into their constitutions. Further, we find only limited evidence of an influence of state supreme court ideology and no evidence for an influence of United States Supreme Court ideology. There appears to be no concern over review/reversal (perhaps because it so rarely occurs), but a strong norm to heed the high court persists, nonetheless. This result is similar to that found at the Courts of Appeals level (Benesh, 2000).
Our results suggest that new institutionalism does not serve well as a general purpose explanation of state supreme court decision making. The state political context matters little. Additionally, the existence of an intermediate appellate court exudes no influence on the behavior of the judges, though criminal jurisdiction does. In short, these well-known institutional constraints are not constraining the judges in our sample of confession cases.
Overall, we do find support for the influence of federal courts on state supreme courts and that influence transcends most all of the known determinants of decision making on the state supreme courts. This leads us to conclude that state supreme courts are faithful agents to their judicial principal (the United States Supreme Court). They are deferential, but do not hesitate to use federalism to their advantage: they can and do decide cases on independent state grounds. In this area of the law, though, they do so without compromising Supreme Court precedent. Other principals, at least in this issue area, are seemingly ignored.
Earlier versions of this article were presented at the 2000 Annual Meeting of the American Political Science Association and Federalism and the Courts: A National Conference, held in Athens, Georgia, February 2001. We wish to thank Matt McCoy and Corina Schulze for their diligent research assistance. We also wish to extend our thanks to Michael Solimine, Harold J. Spaeth, Steve Van Winkle, and Steve Wasby for having read and thoughtfully critiqued earlier versions of this paper. Of course, we bear sole responsibility for any remaining errors and omissions.
1 The most common name states have adopted for their courts of last resort is "supreme court," although not all states use that nomenclature. For simplicity's sake, we use "state supreme court" interchangeably with "state court of last resort" to refer to a state's highest court.
2 463 U.S. 1032, 1040-41 (1983).
3 724 P. 2d 1166 at 1174 (1996).
4 For citations of all Supreme Court confession cases, see Appendix B.
5 This decision rule is arbitrary in that there was no a priori basis for choosing an appropriate cutoff point. However, the distribution of references by the United States Supreme Court makes clear that there is a clear break between those facts referenced in at least a third of the Court's decisions and those referenced less often.
6 Note that for the analysis that follows, we also ran an auxiliary estimation in which we included dummy variables for each of the five case facts, which does not result in any changes in our substantive interpretations.
7 In a very few instances, cases were eliminated because of insufficient information in the text of the court opinion.
8 Cases did not drop out of the original sample at exactly the same rate for each state. However, a compare ison of the original sample of 800 and the final sample of 661 shows a rough proportionality in terms of stratification by state, providing confidence that the final sample is representative of the universe of cases dealing with the voluntariness of confessions.
9 Complete coding rules are available from the authors.
10 We used the following four-point scale: 0 = no relevant state provision, 1= state constitutional protections less than federal constitutional protections, 2 = state constitutional protections equivalent to federal constitutional protections, 3 = state constitutional protections more than federal constitutional protections.
11 For the specifics of logic estimation, see, for example, Aldrich and Nelson, 1984. Further details concerning the model arm available from the authors.
ADMINISTRATIVE OFFICE OF THE UNITED STATES COURTS (2000). Judicial Business of the United States Courts: 1999 Annual Report of the Director. http://www.uscourts.gov/judbus 1999/contents.html.
ALDRICH, J. H., and R D. NELSON (1984). Linear Probability, Logit, and Probit Models. Beverly Hills, CA: Sage Publications.
BAUM, L (1978). "Lower-Court Response to Supreme Court Decisions: Reconsidering 9
a Negative Picture" 3 Justice System Journal 208.
BENESH, S. C. (2002). The U.S. Courts of Appeals and the Law of Confessions: Perspectives on the Hierarchy of Justice. New York: LFB Scholarly Publishing.
- (2000). "Modeling the Supreme Court-Circuit Court Relationship Formally." Paper presented at the Annual Meeting of the Midwest Political Science Association, Chicago.
BERRY, W. D., E. J. RINGQUIST, R. C. FORDING, and R. L. HANSON (1998). "Measuring Citizen and Government Ideology in the American States, 19601993;" 42 American Journal of Political Science 327.
BRACE, P, and M. G. HALL (1995). "Studying Courts Comparatively: The View from the American States," 48 Political Research Quarterly 5.
BRACE, P., M. G. HALL, and L. LANGER (1996). "Judicial Choice and the Politics of Abortion: Institutions, Context, and the Autonomy of Courts:, Paper presented at the Annual Meeting of the American Political Science Association, Washington, D.C.
BRACE, P., L. LANGER, and M. G. HALL (2000). "Measuring the Preferences of State Supreme Court Judges," 62 Journal of Politics 387.
BREHM, J., and S. GATES (1997). Working, Shirking, and Sabotage: Bureaucratic Response to a Democratic Public. Ann Arbor: University of Michigan Press. BRISBIN, R. A., and J. C. KILWEIN (1994). "U.S. Supreme Court Review of State High Court Decisions," 78 Judicature 33.
BUREAU OF JUSTICE STATISTICS (1998). State Court Organization, 1998. http://www.ojp.usdoj.gov/bjs/abstract/sco98.htm. COUNCIL OF STATE GOVERNMENTS (various years). Book of the States.
Lexington, KY: Council of State Governments.
EMMERT, C. F., and C. A. TRAUT (1994). "The California Supreme Court and the Death Penalty," 22 American Politics Quarterly 41.
EPSTEIN, L., and T. G. WALKER (1995). "The Role of the Supreme Court in American Society: Playing the Reconstruction Game," in L. Epstein (ed.), Contemplating Courts. Washington, DC: CQ Press.
EPSTEIN, L., J. A. SEGAL, H. J. SPAETH, and T. G. WALKER (1996). The Supreme Court Compendium: Data, Decisions and Developments, 2nd ed. Washington, DC: CQ Press.
FLEMMING, G. N., D. B. HOLIAN, and S. G. MEZEY (1998). "An Integrated Model of Privacy Decision Making in State Supreme Courts." 26 American Politics Quarterly 35.
GILLMAN, H., and C. W. CLAYTON (1999). "Beyond Judicial Attitudes: Institutional Approaches to Supreme Court Decision-making," in C. W. Clayton and H. Gillman (eds.), Supreme Court Decision-making: New Institutionalist Approaches. Chicago: University of Chicago Press.
GRANO, J. D. (1996). Confessions, Truth, and the Law. Ann Arbor: University of Michigan Press.
HALL, M. G. (1992). "Electoral Politics and Strategic Voting in State Supreme Courts:' 54 Journal of Politics 427.
HALL, M. G., and P. BRACE (1999). "State Supreme Courts and Their Environments: Avenues to General Theories of Judicial Choice," in C. W. Clayton and H. Gillman (eds.), Supreme Court Decision-making: New Institutionalist Approaches. Chicago: University of Chicago Press.
(1996). "Justices' Responses to Case Facts: An Interactive Model," 24 American Politics Quarterly 237.
- (1994). `The Vicissitudes of Death by Decree: Forces Influencing Capital Punishment Decision-making in State Supreme Courts" 75 Social Science Quarterly 136.
- (1992). "Toward an Integrated Model of Judicial Voting Behavior," 20 American Politics Quarterly 147.
INBAU, F. E., and J. E. REID (1962). Criminal Interrogation and Confessions. Baltimore, MD: Williams and Wilkins Company.
KORT, F. (1963). "Content Analysis of Judicial Opinions and Rules of Law; in G. A. Schubert (ed.), Judicial Decision-making. Glencoe, IL: Free Press.
LANGER, L. (1999). "Strategic Behavior and Policy Saliency to the Elected Elite: A Comparative Examination of State Judicial Review Across Four Areas of Law, 1970-1993:' Paper presented at the Annual Meeting of the Midwest Political Science Association, Chicago.
- (1997). "State Supreme Courts and Countermajoritarian Behavior: Strategic or Sincere?" Paper presented at the Conference on the Scientific Study of Judicial Politics, Emory University, Atlanta.
MARTINEK, W. L. (2000). "Judicial Impact: The Relationship between the United States Supreme Court and State Courts of Last Resort in Search and Seizure Decision Making." Ph.D. diss., Michigan State University, East Lansing.
MITNICK, B. M. (1991). "An Incentive Systems Model of the Regulatory Environment" in M. J. Dubnik and A. R. Gitelson (eds.), Public Policy and Economic Institutions. Greenwich, CT: JAI Press.
- (1980). The Political Economy of Regulation: Creating, Designing, and Removing Regulatory Forms. New York: Columbia University Press.
- (1975). "The Theory of Agency: The Policing 'Paradox' and Regulatory Behavior," 24 Public Choice 27.
MOE, T. M. (1987). "An Assessment of the Positive Theory of 'Congressional Dominance,"' 12 Legislative Studies Quarterly 475.
NATIONAL CENTER FOR STATE COURTS (1995). State Court Caseload Statistics, 1994. Williamsburg, VA: National Center for State Courts.
REDDICK, M. (1997). "The Applicability of the Legal and Attitudinal Models to the Treatment of Precedent in the Courts of Appeals" Ph.D. diss., Michigan State University, East Lansing.
ROHDE, D. W., and H. J. SPAETH (1976). Supreme Court Decision Making. San Francisco: W. H. Freeman.
ROSENBERG, G. N. (1991). The Hollow Hope: Can Courts Bring About Social Change? Chicago: University of Chicago Press.
ROSS, S. A. (1973). "The Economic Theory of Agency: The Principal's Problem:' 62 American Economic Review 134.
RUTLEDGE, D. (1994). Criminal Interrogation: Law and Tactics. Placerville, CA: Copperhouse Publishing Company.
SCHAFER, W. J. III (1968). Confessions and Statements. Springfield, IL: Charles C. Thomas, Publisher.
SEGAL, J. A. (1997). "Separation-of-Powers Games in the Positive Theory of Congress and Courts:' 91 American Political Science Review 28.
SEGAL, J. A., and A. D. COVER (1989). "Ideological Values and the Votes of U.S. Supreme Court Justices:' 83 American Political Science Review 557.
SEGAL, J. A., L. EPSTEIN, C. M. CAMERON, and H. J. SPAETH (1995). "Ideological Values and the Votes of U.S. Supreme Court Justices Revisited," 57 Journal of Politics 812.
SEGAL, J. A., and H.. J. SPAETH (1993). The Supreme Court and the Attitudinal Model. Cambridge, MA: Cambridge University Press.
SONGER, D. R. (2000). "Partisan Preferences, Contextual Influences, and the Impact of Changing Supreme Court Policy in High and Low Salience Cases:' Paper presented at the Annual Meeting of the Midwest Political Science Association, Chicago.
SONGER, D. R., and S. HAIRE (1992). "Integrating Alternative Approaches to the Study of Judicial Voting: Obscenity Cases in the United States Court of Appeals," 36 American Journal of Political Science 963.
SONGER. D. R., J. A. SEGAL, and C. M. CAMERON (1994). "The Hierarchy of Justice: Testing a Principal-Agent Model of Supreme Court-Circuit Court Interactions:' 38 American Journal of Political Science 673.
TRAUT, C. A., and C. R EMMERT (1998). "Expanding the Integrated Model of Judicial Decision Making: The California Justices and Capital Punishment," 60 Journal of Politics 1166.…
Questia, a part of Gale, Cengage Learning. www.questia.com
Publication information: Article title: State Supreme Court Decision Making in Confession Cases. Contributors: Benesh, Sara C. - Author, Martinek, Wendy L. - Author. Journal title: Justice System Journal. Volume: 23. Issue: 1 Publication date: January 1, 2002. Page number: 109+. © National Center for State Courts 2008. Provided by ProQuest LLC. All Rights Reserved.
This material is protected by copyright and, with the exception of fair use, may not be further copied, distributed or transmitted in any form or by any means.