The U.S. Supreme Court is but one of three political institutions within the structure of the U.S. federal government. Within this system of separated powers it rules on the constitutionality of some of the nation’s most important legal and political issues. In making such decisions, the nation’s highest court may be considered the most powerful of the three branches of the U.S. federal government. Understanding this process will allow scholars, students of the Court, and Court watchers alike to gain a better understanding of the way in which the justices conduct their business and to come to terms with some of the most important legal and political decisions in our nation’s history. Combining a theoretical account of Supreme Court decision-making with an examination of its internal decision-making process illuminates this opaque institution.
The U.S. Supreme Court is but one of three political institutions within the structure of the U.S. federal government. Within this system of separated powers it rules on the constitutionality of some of the nation’s most important legal and political issues. 1 Since the turn of the 21st century alone, the Court has made decisions that affected the outcome of a presidential election (Bush v. Gore), universal health care (National Federation of Independent Business v. Sebelius), the liberty of same sex couples (Lawrence v. Texas and Obergefell v. Hodges), and the voting rights of minorities (Shelby County v. Holder). In making such decisions, the nation’s highest court may be considered the most powerful of the three branches of the U.S. federal government. 2
This may be a controversial position given the powers held by the elected branches at the federal level. Indeed, Congress has clear and important powers explicated in Article I of the Constitution. As such, among other powers, it declares war, decides how to raise and spend money, and ratifies all international treaties. At the same time, the president is the nation’s chief executive and commander in chief of the military. In short, while the U.S. Supreme Court hears and decides only about 75 cases per term, Congress and the executive branch wield their powers on a daily basis. However, given the issues on which the Court sets legal policy, it is not a stretch to suggest that the justices do indeed wield a great deal of power.
Beyond the debate about the power of each branch, the two elected branches often carry out their jobs in the public eye. Indeed, the president is the most visible political figure in the nation, and the work of Congress is covered (at a minimum) on a regular basis by the mass media. In fact, three cable television channels and a radio network (C-SPAN, C-SPAN2, C-SPAN3, and C-SPAN Radio) devote themselves to allowing the public to watch floor debates and votes as well as virtually all committee proceedings. In short, the elected branches enjoy clear power that is often conducted transparently on a daily basis.
In contrast, with the exception of one hour set aside for litigants to present oral arguments in most cases it decides, and the public announcement of its decisions, the work of the U.S. Supreme Court is conducted almost completely outside of the public’s eye. 3 As a result, the Court’s decision-making process is largely opaque, and therefore the public knows very little about how the justices reach the decisions that affect every part and every citizen of the United States. An attempt is made to illuminate this process so that scholars, students of the Court, and Court watchers alike can gain a better understanding of the way in which the justices conduct their business and come to terms with some of the most important legal and political decisions in our nation. The conventional theoretical account of Supreme Court decision-making is briefly overviewed and the Court’s internal decision-making process from agenda setting to opinion writing is examined.
Supreme Court justices are policy-oriented strategic decision makers (Epstein & Knight, 1998; Eskridge, 1991a, 1991b; Ferejohn & Weingast, 1992; Gely & Spiller, 1990), which means their decisions are constrained by a host of factors (Maltzman, Spriggs, & Wahlbeck, 2000; Epstein & Knight, 1998; Johnson, 2004; Black & Owens, 2012; Black, Wedeking, & Johnson, 2012). That is, when making decisions, they must account for the preferences of their immediate colleagues, the preferences of actors beyond the Court, and institutional norms and rules that might affect the decisions that they can make. The three prongs of this model are considered.
That policy is the main goal of Supreme Court justices is neither a new nor a controversial idea. Rather, this argument is well grounded in the work of legal realists such as Llewellyn (1931) and Frank (1949), and early judicial behavior scholars such as Pritchett (1948), Murphy (1964), and Schubert (1965) and Segal and Spaeth (2002). Scholars have provided empirical support for this argument in several ways. First, individual justices’ voting patterns are very consistent over time. For instance, with the exception of two terms ( 1974 and 1977 ) Lewis Powell voted liberally in civil liberties cases no more than 43% of the time in any given term. Likewise, William Brennan’s liberal support for civil liberties fell below 70% only one term during his Court tenure ( 1969 ) (Epstein, Segal, Spaeth, & Walker, 2011, p. 456). This consistency indicates justices pursue specific policy goals, and rarely waver from doing so.
Beyond voting patterns, Johnson (2004) indicates that the vast majority of questions justices ask during oral arguments concern policy. After these proceedings, Epstein and Knight (1998, pp. 30–32) demonstrate almost 50% of all remarks made by justices during the Court’s conference discussions concern policy and 65% of statements in circulating memoranda during the opinion-writing process address policy considerations. These remarks include statements about legal principles the Court should adopt, courses of action the Court should take, or a justice’s beliefs about the content of public policy. Finally, scholars address the interactions that take place between justices during the opinion-writing process (Maltzman, Spriggs, & Wahlbeck, 2000; Epstein & Knight, 1998; Murphy, 1964). They point to justices’ bargaining statements during the opinion-writing phase of a case to demonstrate that policy considerations are the driving force behind justices’ decisions.
The attitudinal model of Supreme Court decision-making suggests that justices are unconstrained in their ability to vote for their most preferred policy outcomes because they enjoy life tenure (Segal & Spaeth, 2002). In other words, because justices do not face election or retention, and because they usually do not have higher political ambitions, they can vote for their most preferred outcomes without consequence. In contrast, the strategic model suggests that, although they pursue policy goals, justices cannot always make decisions that conform perfectly to their preferences. Rather, because five justices must usually agree on a decision to set precedent justices must pay particular attention to the preferences, and likely actions, of their immediate colleagues. In short, Supreme Court justices alter their behavior in order to achieve their goals within the context of making decisions by majority rule.
A recent, yet rich, literature explores the extent and impact of internal bargaining between justices (see, e.g., Ringsmuth, Bryan, & Johnson, 2013; Johnson, Spriggs, & Wahlbeck, 2005; Maltzman, Spriggs, & Wahlbeck, 2000; Caldeira, Wright, & Zorn, 1999; Epstein & Knight, 1998). These works are progeny of Murphy (1964), who argued that justices are rational actors and act as such when deciding cases. The reason for this is obvious, as Murphy notes: “Since he shares decision making authority with eight other judges, the first problem that a policy oriented justice would confront is that of obtaining at least four, and hopefully eight, additional votes for the results he wants and the kinds of opinions he thinks should be written in cases important to his objectives” ( 1964 , p. 37).
While Murphy did not systematically test his theory, others have done so. For example, in an analysis of Justice Brennan’s and Justice Marshall’s private papers, Epstein and Knight (1995) demonstrate that over 50% of cases in one sample contained one or more bargaining statements between the justices. 5 In a later monograph, Epstein and Knight conclude that, “law, as it is generated by the Supreme Court, is the result of short-term strategic interactions among the justices and between the Court and other branches of government” ( 1998 , p. 18).
Wahlbeck, Spriggs, and Maltzman (1998) support these findings in their empirical analysis of opinion circulation on the Court. They find that an opinion goes through more drafts as the ideological heterogeneity of a majority coalition increases, as the number of suggestions given to the opinion writer by other justices increases, as the number of threats made to the opinion writer increases, and as the number of times other justices say they are yet unable to join an opinion increases. This suggests to Wahlbeck et al. that, “Opinion authors’ actions are shaped by the interplay of their own policy preferences and the actions of their colleagues” (p. 312).
Wahlbeck and his colleagues also find evidence that the decision to join a majority opinion is a strategic choice as well ( 1998 ). Specifically, they demonstrate that the decision to join is determined by how acceptable a majority opinion is to a specific justice, whether that justice can attain concessions from the opinion writer, and the past relationship between the opinion writer and the justice deciding whether to join. Finally, Maltzman, Spriggs, and Wahlbeck (2000) provide evidence that how the chief justice assigns opinions, how justices respond to initial opinion drafts, and how coalitions form are all processes grounded in strategic interaction. 6 This means that the process through which the Court makes decisions is a product of interactions and interdependencies between the justices. If, on the other hand, justices simply voted for their most preferred outcomes, there would be no evidence of bargaining and accommodation behind the scenes of the decision-making process.
More recently Black, Schutte, and Johnson (2013) and Johnson et al. (2005) demonstrate that justices use the rules of the game in a strategic manner. The former analyze how a justice can use threshold issues to keep the Court from deciding a case far from her preferred outcome. 7 In addition, Johnson and his colleagues find that chief justices (as well as senior associate justices) can and do manipulate the voting rules during the Court’s conference discussions to move a decision closer to their preferred outcomes. Both of these recent works extend and enhance empirically the theoretical concept that justices are strategic political actors.
For instance, Knight and Epstein (1996) argue that justices adhere to the norm of respecting precedent. While their findings are far from general (they analyze only 13 cases), the evidence is nonetheless compelling. Indeed, if respect for precedent were not a norm, then Knight and Epstein would not have found evidence that the justices frequently discuss past cases in their private deliberations. That the justices make such references to precedents in private memos suggests that they act as if they, themselves, are constrained to follow these decisions. The question, however, is why do the justices feel constrained by precedent? For Knight and Epstein the answer is simple: “compliance with this norm is necessary to maintain the fundamental legitimacy of the Supreme Court” ( 1996 , p. 1029). In other words, they argue that if the Court frequently ignored its own legal precedents its credibility as a judicial institution might be questioned, and it could potentially lose legitimacy—its main source of power.
Respecting precedent is an informal norm, but the Court must also follow certain formal rules such as those set out in the Constitution. Because the Constitution gives Congress the power to override Supreme Court decisions, the justices must account for the preferences of Congress when deciding where to set policy in a particular area of law (see, e.g., Clark, 2009; Ringsmuth & Johnson, 2013). Other codified rules are found in Article III of the Constitution; these include the Court’s jurisdiction to hear certain cases, 8 the requirement that a party must have standing (Flast v. Cohen [1968]) to be heard in the Supreme Court, and that a case must be justiciable before the Court will consider ruling on it. 9
Four key aspects of the Court’s decision-making process are considered: agenda setting, oral arguments, conference, and opinion writing. Each demonstrates quite clearly that justices on our nation’s highest court are strategic actors.
The U.S. Supreme Court is a passive institution and must wait for cases to come to it. In other words, the justices cannot introduce legislation like members of Congress. 10 However, as strategic decision makers, justices can do encourage litigants to submit cases presenting a particular question or issue area (see, e.g., Hausseger & Baum, 1999). The process through which cases are placed on the Court’s docket is briefly discussed. Research that focuses on the key rule that governs the Court’s agenda-setting process—the Rule of Four—is presented.
A party who loses in a lower federal court or in a state supreme court (if the case involves a federal question) may appeal to the U.S. Supreme Court. 11 When the appeal, or petition, is submitted, the clerk of the Court provides notice to the respondent (i.e., the winner in the lower court). The respondent has the option to respond, waive right of response, or do nothing at all. Most respondents choose the first option, and therefore both parties file briefed (written) legal arguments as to why the Court should or should not hear the case. 12 These briefs are then distributed to the justices’ chambers for review. Over time there has been a clear increase in the number of petitions submitted to the Court. In fact, there has been a five-fold increase in the number of petitions submitted between 1935 and 2015 . At the same time, while the number of cases granted review increased dramatically through the late 1970s this number has fallen just as precipitously between then and the present day.
Once the parties submit their cert. briefs, the first hurdle to winning review by the Court is making the “discuss list.” This list includes all petitions for review the justices will discuss and formally vote on during their weekly conference meetings. The chief justice creates the first draft of the discuss list. 13 Any associate justice can add a petition to the chief’s list but he or she may not remove a case already on the list. 14 While there are little data on the Court’s agenda-setting (or cert. process) the justices do discuss a large number of possible cases each term. For instance, over eight terms between 1986 and 1993 the Court discussed approximately 800 petitions per term (Black & Boyd, 2013).
Conference voting at the cert. stage is, like voting on case outcomes, sequential. The justice who places a case on the discuss list speaks first and usually offers a justification for why a petition should be granted review. From there the nine vote on whether to review the case. They do so in descending order of seniority with the chief justice considered the most senior justice. It takes four votes to grant review and if four votes are not forthcoming the Court denies the cert. petition. In other words, the Court will not hear the case. Ultimately, this means the lower court decision remains the law. That said, such a denial does not mean the justices necessarily agree with the lower court decision. Rather it simply means they could not agree to hear the present case.
As noted, it takes four votes for the Court to hear a case. This important institutional rule is unique because it allows a minority of justices to both set the Court’s agenda and to change, rather than simply preserve, the status quo. 15 That is, by granting a hearing and by then issuing a ruling on a case from a lower court, the Supreme Court sets national doctrine by either applying the lower court’s ruling to the entire country or by reversing the ruling of the lower court altogether. This is an important power for two main reasons. First, it acts as a sharp constraint on majority tyranny at the Court’s agenda-setting stage. As Kurland and Hutchinson (1983, p. 645) put it, “The rule of four is a device which a minority of the Court can impose on the majority a question that the majority does not think it appropriate to address.” The potency of this rule is not lost on the justices. As Justice Brennan (1973) put it, choosing cases is “second to none in importance.” It also clearly worries at least one former justice. As John Paul Stevens (1983, p. 19) pointed out: “Every case that is granted on the basis of four votes is a case that five members of the Court thought should not be granted. For the most significant work of the Court, it is assumed that the collective judgment of its majority is more reliable than the views of the minority.”
The historical record on the Rule of Four is incomplete (Stevens, 1983; Revesz & Karlan, 1988; O’Brien, 1997; Epstein & Knight, 1998; Hartnett, 2000). We know, however, that its origins come sometime after passage of the Evarts Act of 1891 . This law established the circuit courts of appeals and codified that no right of appeal to the Supreme Court existed. The result was that the justices had much greater discretion over their appellate docket. As Hartnett (2000) put it, “thus was born the then revolutionary, but now familiar, principle of discretionary review of federal judgments on writ of certiorari.” Although there is evidence justices relied on a minority certiorari rule through the late 1800s and early 1900s, it was not until 1925 that its use became public when Justice Willis Van Devanter appeared before the House Judiciary Committee during its hearings on the Judges’ Bill. 16
Van Devanter’s purpose was to “assure Congress that increased control over its [the Court’s] own docket would not lead to arbitrary dismissal of cases” (Robbins, 2002, p. 12).
More specifically, to assuage the worry that the Court would reject cases that could be potentially important, Van Devanter explained that:
We always grant petitions when as many as four think that it should be granted and sometimes when as many as three think that way. We proceed upon the theory that, if that number out of nine are impressed with the thought that the case is one that ought to be heard and decided by us, the petition should be granted.
( 2002 , p. 12)A decade later, Chief Justice Hughes reiterated Justice Van Devanter’s response to the congressional concern that the Court may not take cases important for the law because of the justices’ discretion over their docket. In a speech before the American Law Institute he noted, “we are liberal in the application of our rules and certiorari is always granted if four justices think it should be, and, not infrequently, when three, or even two, justices strongly urge the grant” (Hughes, 1937, p. 459). The point is that for at least the past 80 years a minority of the justices has controlled the Court’s agenda.
Empirical work on the Rule of Four focuses almost exclusively on how it affects the size of the Court’s docket each term. For instance, Stevens (1983) argues that the Rule of Four comes into play in about 25% of all cases that make the discuss list. He concludes that many of these cases are probably unimportant, and should therefore be left off of the plenary docket. O’Brien (1997) obtains similar results in his analysis of Justice Marshall’s docket books for the 1990 term. He finds that 22% of cases decided during this term were granted certiorari with only four votes.
Perry and Carmichael (1985–1986) take the question of case selection a bit further. They test whether the Rule of Four protects “important” cases. By their operationalization this does not happen because most important cases almost always receive at least five votes for certiorari. Perry and Carmichael point out, however, that if the Court is interested in taking “nearly significant” cases it should not abandon its long-lasting rule.
While Perry and Carmichael suggest the Rule of Four protects somewhat important cases, the normative implication of Stevens’s and O’Brien’s findings is that the Court should consider abandoning this rule. For Stevens, the quarter of all cases docketed with fewer than five votes presented an additional and unnecessary burden on him and his colleagues. Indeed, Stevens believes the Court should decide only the most important cases, and therefore the problem of overworked justices could be abated by taking only those cases with a majority vote on certiorari.
Beyond the debate between legal scholars and justices, the Rule of Four has drawn scorn from the mass media as its incompatibility with majority rule has come to light in death penalty cases (Liptak, 2007). A prisoner sentenced to death needs the vote of a simple majority, or five justices, to stay or postpone his or her execution, yet the Rule of Four allows a minority of justices to place a prisoner’s appeal on the docket. This sets up the possibility that the Court could simultaneously grant a prisoner’s petition to appeal his or her sentence while refusing to stay the execution that would, in the legal lexicon, “moot” the case if the prisoner was subsequently executed.
Certainly the normative implications of the Rule of Four are interesting, but this line of work fails to address a fundamental question: Why would a minority coalition want to place a case on the docket when five of their colleagues could either vote to dismiss the case as improvidently granted (DIG) at the plenary stage, or simply outvote them at the merits stage? 17 After all, on the surface the Rule of Four is incompatible with the rule that a simple majority of justices can vote to dismiss. As such, a preference cycle could exist whereby a case was continually granted and then dismissed (Riker, 1982). Two explanations have been given in the literature for why this does not happen on the Court. Regarding DIGs, Epstein and Knight (1998, p. 120) suggest a norm exists whereby the five justices who voted against certiorari cannot form the five-member coalition to DIG a case. While these scholars point out that this norm can be and has been violated, justices do not often do so. The result is that Rule of Four cases ultimately receive treatment similar to cases granted review with five or more votes.
With respect to the latter point, scholars have offered some answers, albeit not theoretically satisfying ones. For instance, in her analysis of case selection based on Justice Burton’s docket sheets Provine (1980, p. 157) finds, “that the desire to be agreeable and the leadership responsibility felt by chief justices are the primary reasons some justices vote oftener for review in four-vote cases than otherwise.” She therefore concludes that, “The hypothesis that four-vote cases reflect the presence of coalitions seeking review on the merits receives no support in this analysis” (Provine, 1980, p. 158). This conclusion is based on the fact that the two most frequent members of four-vote certiorari coalitions were Justices Burton and Clark, both of whom were considered “affable and outgoing in their personal relationships” (p. 156). The point for Provine (1980) is that there seems to be nothing strategic about Rule of Four cases, and that the key explanation for justices joining these minority coalitions comes from a sense of friendship, from wanting to be deferential to their colleagues, or from a desire to lead the Court fairly (for chief justices).
The analysis provided by legal scholars is both theoretically and empirically unsatisfying. As such, several political scientists have attempted to systematically analyze the Rule of Four. In his seminal work on Supreme Court agenda setting, Perry (1991) argues there are times when justices engage in strategic behavior during the certiorari stage, and the Rule of Four may encourage such behavior. Perry (1991, p. 98) also provides evidence that there are times when a coalition of four will not force a case onto the docket because the justices in that coalition know they will surely lose on the merits strategy known as a defensive denial.
Epstein and Knight (1998) go a step further than Perry by providing convincing evidence to support the argument that the Rule of Four can be used for strategic purposes. As they point out ( 1998 ), “The Rule of 4 invites forward thinking. Policy oriented justices know that if they are to attain their goals they must take those cases they believe will lead to their preferred outcomes and reject those that will not.” The key for them, then, is that justices can use this rule to make “strategic calculations throughout the decision making process” (p. 121).
In the end, this part of the Court’s decision-making process plays a vital role for the justices. Indeed, it is clear that decisions made at this stage of the process have a great deal to do with the decisions the justices make on the merits of a case.
Litigants lucky enough to have a case accepted for review by the Supreme Court submit a second round of briefs. These briefs are meant to convince the justices who should win the case. The rules surrounding these briefs on the merits are discussed along with the research that seeks to explain the extent to which written arguments affect the decisions justices make.
“Rules of the Supreme Court of the United States” is an 83-page document created by the Court that provides, among other things, several sections of detailed information on how to properly file, format, and write a brief. The real paper-shuffling begins once the Court has accepted a case: from this point the petitioner has 45 days to file a merits brief, and the respondent’s brief is required 30 days later. Petitioners are allowed a reply to the respondent’s brief, but it must be submitted at least one full week before oral arguments are heard, and no more than 30 days after the respondent’s brief is filed. Litigants may request extensions on these time frames; however, “an application to extend the time to file a brief on the merits is not favored” (Rule 25.5).
In addition to a rigid time frame, the Court also outlines specifics for the formatting and length of briefs. As dictated by Rule 33, all briefs (merits, response, amicus) are printed on 60-pound paper in booklet format measuring 6 1/8th inches by 9 1/4th inches. These booklets are then bound by saddle stitch or perfect binding, and covered with 65-pound card stock paper of the appropriate color. Merits briefs for both petitioner and respondent are limited to 15,000 words and covered in light blue or red paper, respectively. The petitioner’s reply brief is not to exceed 6,000 words and should have a yellow cover. Each brief may include appendixes of unlimited length, however additional arguments should not appear in these appendixes. 18
One source of briefs not yet discussed is amicus groups. Amici, or friends of the Court, are most often interest groups or established organizations that file a brief in support of either side. 19 Such briefs are limited to 9,000 words, and should “bring to the attention of the Court relevant matter[s] not already brought to its attention by the parties” (Rule 37.1). Although limited in words, there is no limit to the number of separate amicus groups that can file. In the 2012 term, for example, multiple amicus briefs were filed in 67 of the 76 cases granted review. Amicus participation ranged from zero to 97 briefs—the former occurring in only two cases, with the latter in support of Hollingsworth v. Perry (2013), one of the term’s two same-sex marriage cases.
Finally, Supreme Court rules dictate that forty copies of every single brief (petitioner, respondent, reply, and amicus) are delivered to the Court in hard copy, as per the rules explained previously. Considering the hundreds—if not thousands—of pages filed for every case appearing before the Court, such strict rules are necessary to maintain any semblance of order and organization. With so much time, effort, and paper spent on brief writing—and subsequent reading—the extent to which briefs affect outcomes is considered.
Litigant and amicus briefs serve two broad purposes: to provide information to the Court, and to set the bounds of policy space from which legal opinions can be drawn. The informational nature of briefs is evident given the pure volume provided to the Court, but what sort of information is contained in these myriad pages? Johnson (2004) offers such insight; he provides a breakdown of issues briefed in a random sample of 75 Burger Court civil liberties cases. Note that litigant briefs most often address policy and constitutional issues—this is not surprising given the description of justices as seekers of policy preferences (Epstein & Knight, 1998; Segal & Spaeth, 2002). Indeed, these two issues dominate litigant briefs and account for over 70% of all briefed issues.
Further, when both amicus and litigant participation are considered, the Court still receives the most information regarding policy and constitutional issues. Note, however, there are clear differences in the types of information each brief provides. Specifically, amicus briefs refer more often to precedent and external actors. This is, again, unsurprising as amici are by definition an external actor; it would make sense that their briefs would inform the Court on their preferences as well as additional precedent that would support their preferences.
Next, existing research demonstrates that legal briefs submitted to the Court often set the boundaries of a case by framing the issues for the justices (Wahlbeck, 1998). To this end, the legal model of decision-making posits that justices care about the law and are bound to it, and scholarly work in this area agrees. For example, Epstein and Kobylka (1992) examine legal change in the jurisprudence of abortion and death penalty cases. While they consider factors such as public opinion and a changing political and social environment, Epstein and Kobylka ultimately conclude, “it is the law and legal arguments as framed by legal actors that most clearly influence the content and direction of legal change” ( 1992 , p. 8). Later work by Epstein, Segal, and Johnson (1996) examines briefs and opinions, concluding that the Court follows the doctrine of sua sponte—a norm disfavoring issue creation during opinion writing. Finally, Corley (2008) uses plagiarism software to detect when majority opinions “borrow” phrases and sentences from litigant briefs. She finds the quality of a brief—as defined by the experience of the writing attorney—positively affects the chances a justice will “borrow.” Taken together, this line of research suggests that briefs directly impact the decision-making process by providing the Court with valuable information, and by constraining their choice-set by setting policy boundaries.
Once the Supreme Court grants review in a case the parties file legal briefs to convince the justices how to decide. In addition to the briefs, the justices sit for oral arguments. During these proceedings attorneys for each side of a case present their best arguments to the justices in an effort to convince the Court to rule in a particular way. This section considers the role justices play in this part of their decision-making process.
In the Supreme Court’s early days great lawyers such as Daniel Webster, John Calhoun, William Pinkney, and Henry Clay often appeared before the justices. In this time period oral arguments were elaborate oratories but, more important, they often provided the justices with their only source of information about a case: briefs were rarely if ever submitted and outside parties did not submit amicus curiae (friend of the Court) briefs as they do today. The result was that the justices placed no time limitation on the argument sessions. As such, advocates sometimes spoke for many hours over multiple days. In one case, for instance, Davis (1940) points out that Webster and his rival argued for a full ten days. In stark contrast to contemporary arguments, historians suggest that the justices rarely interrupted the advocates with questions or comments (compare Warren, 1922 with Johnson et al., 2009).
Certainly, hearing arguments over many days was possible because the Court heard so few cases in its early days. However, the Court’s rising caseload soon made such indulgences impossible. In addition, it seems that, at some point, the justices could no longer handle such long sessions. As one biographer put it, Chief Justice John Marshall complained of boredom (Beveridge, 1929) and Justice Joseph Story found the arguments “excessively prolix and tedious” (Hughes, 1928). Just as certainly as days-long arguments were possible for the early Court, such a model was untenable as the justices’ caseload increased. As a result, in 1849 they instituted Rule 53, which limited each attorney to a two-hour argument (Frankfurter & Landis, 1928). The justices simultaneously required the first written arguments, consisting of an abstract of points and authorities (Frankfurter & Landis, 1928). Today, and since 1970 , the time allotted for these has been limited to 30 minutes per side.
The modern time constraints on oral arguments may be due to the fact that the justices have so much information at their disposal prior to these open court sessions. Indeed, today they possess litigant briefs (Epstein & Kobylka, 1992), briefs amicus curiae (Spriggs & Wahlbeck, 1997), briefs on certiorari (Caldeira & Wright, 1988), media accounts (Epstein & Knight, 1998), and lower court opinions. Further, unlike the Court’s early days, when the justices were transfixed by the great orators (or put to sleep from boredom) they largely control the argument sessions today. Specifically, one analysis of 347 cases over four recent terms included 43,000 utterances and 1.4 million words spoken by the justices (Black et al., 2009). It seems then, in modern cases, the justices now speak as much or more so than do the attorneys. The question is whether such questioning has some bearing on how the justices decide.
As with the other aspects of the decision-making process, evidence accumulated over the last decade establishes that, generally, oral arguments play an integral role in the Court’s decision-making process (Johnson, 2004; Wrightsman, 2008). First, as policy-oriented political actors, justices are clearly concerned with questions of policy. Johnson (2004) demonstrates that in cases without amicus participation, 40% of the Court’s questions focus on policy, and this increases to 43% when amici participate. Second, justices spend a great deal of time asking questions about the preferences of actors external to the Court (Johnson, 2004). Indeed, in his sample of cases Johnson (2004) shows that 1,159 (36%) questions fall into this category when amici are not present in a case. This total is similar when amici participate: 787 (34%). Finally, justices raise questions about institutional constraints (precedent and threshold issues) they may face, but fewer of them than about the other issues. Specifically, when amici participate, 13.7% of all questions focus on precedent or threshold issues, and 13.2% focus on these issues in cases without amici.
Beyond the information they provide to the justices, there is evidence that the quality of oral arguments forwarded by attorneys during these proceedings affects justices’ votes (Johnson, Wahlbeck, & Spriggs, 2006). Indeed, even justices predisposed to vote for a particular side (based on their ideological predilections) tend to vote more often for the side that offers better arguments in open Court. Finally, there is mounting evidence that, during oral arguments, justices foreshadow how they will decide (Shullman, 2004; Roberts, 2005; Johnson, Black, Goldman, & Treul, 2009; Black, Wedeking, & Johnson, 2012). Such signals emanate from the number of questions justices ask the attorney for each side of the dispute as well as from the emotive tenor of these questions. When justices give one side a harder time (by asking more questions) or when they ask that side questions using less pleasant language, it is more likely to lose the case (while this does not necessarily show causality the correlation between these phenomena is quite robust).
Justices’ conversations with attorneys are not the only discussions that take place during oral arguments. Rather, they are prone to speak to one another as well. Anecdotally, Wasby et al. argue that, “it is not surprising that the judges would use part of the oral argument time for getting across obliquely to their colleagues on the bench arguments regarding the eventual disposition of a case” (Wasby, D’Amato, & Metrailer, 1977, p. xviii). They conclude elsewhere that, “Another, less noticed function is that oral argument serves as a means of communication between judges” (Wasby, D’Amato, & Metrailer, 1976, p. 418). Recent work corroborates these anecdotal findings. Black et al. (2012) provide systematic evidence that justices speak to (and often speak over) one another, listen to their colleagues’ questions and comments, and use the oral arguments to predict the outcome of the case.
Overall, despite the conventional wisdom through the last decade of the 20th century, scholars have now made clear that the hour-long sessions in open Court can and do affect the decisions justices make. They clearly elicit information from the attorneys that helps them do so. In addition, they make their positions clear with how they ask questions and also speak to each other through their questions and comments. What happens once the justices retire back to the confines of the Marble Palace?
Chief Justice Roberts has said publicly that, “We [the Court] are the most transparent branch of government. Everything we do that has an impact is done in public. . . . You see our work in public at the Court. Our decisions are out there” 20 Interestingly, the chief’s contention is the fact that, unlike the elected branches, the Court issues explicit public justifications (in the form of written opinions) for the decisions they make. The problem with this argument is that the public aspect to which the chief refers is only the end product. With the exception of the oral arguments (see Decisions on the Merits: Oral Arguments) the remainder of the decision-making process occurs behind closed doors in the marble palace. The justices’ conference discussions and why these discussions are important for how justices decide are examined.
While the justices sit for oral arguments on Mondays, Tuesdays, and Wednesdays during the term, private conference discussions take place on Wednesdays and Fridays. These discussions serve two purposes—to choose cases from the discuss list to set for future arguments and to vote on the cases that have already been argued.
Because the justices meet in conference twice a week, they use Wednesday conferences to discuss the merits (outcomes) of cases heard Mondays and Tuesdays. Friday conference is therefore reserved for cases argued on Wednesdays. 21 These meetings, held in the conference room that adjoins the chief’s chambers, are completely private. That is, nobody is allowed in the room except the justices. If something is needed from outside the room the most junior justice calls to make the request. 22
When conference begins the chief justice presents the facts of the first case. From there he offers his personal view of the case and then casts his vote. 23 When the chief is done the associate justices offer their views and votes in order of seniority. That is, the most senior associate justice presents and votes next with each justice doing so until the Court’s newest member finishes the discussion. 24 Usually the justices have a fairly good idea of how they will vote in the case, but there are times when they are less certain. 25 The justices repeat this process for each case argued during the current week.
Scholars have emphasized for decades that conference votes are only the tip of the iceberg for the business the justices conduct (see, e.g., Epstein & Knight, 1998). However, the only data normally available to scholars are the conference votes. 26 Little information exists about the legal and policy issues the justices discuss during conference, which means scholars have largely ignored this part of the Court’s decision-making process. Despite the paucity of research in this area there is evidence to suggest the justices discuss the key aspects of cases they decide. Johnson (2004) demonstrates that, during conference, the justices clearly pick up on issues briefed by the parties and on issues discussed during oral arguments. More specifically, justices discuss policy options and key precedents during their private conference discussions. Similarly, Knight and Epstein (1996) find that the justices clearly discuss precedent at conference. For them, this suggests the justices believe they are bound by the norm of respecting past decisions. In short, while scholars do not yet have a full picture of what transpires during conference, these studies provide insights that have, until the past decades, eluded scholars and Court watchers alike.
Beyond the discussions about the specific issues the justices must decide, conference is important because how the justices vote determines who will ultimately write the majority opinion in a case. If the chief justice votes with the majority at the end of a case discussion he has the authority to choose who write the majority opinion. 27 This prerogative helps him influence the Court’s agenda because he can select either an author whose opinion is close to his own preferences or one who will minimize the prospective policy loss if the chief’s preferred outcome does not prevail (Epstein & Knight, 1998; Maltzman, Spriggs, & Wahlbeck, 2000). More specifically, Maltzman, Spriggs, and Wahlbeck (2000) demonstrate that the justice who assigns the opinion has some power to set the agenda for the majority coalition because the writer gets the first move in the bargaining process. The chief exercises this discretion and guides the opinion toward his preferred position by assigning it to ideologically proximate justices (or to himself), especially in important cases. In other words, although the chief justice is constrained by norms, like equitable distribution of assignments, he has the power to choose who will articulate the Court’s opinion and the lens through which a case will be decided. 28
There is clear evidence that chiefs (and sometimes senior associates) act strategically to ensure the opinion assignment power. Johnson, Spriggs, and Wahlbeck (2005) note that Chief Justice Warren Burger was renowned for casting votes at conference that would allow him to control the Court’s agenda through opinion assignment. Indeed, Burger often changed his initial votes to join the majority coalition, cast “phony votes” by voting against his preferred position, and sometimes declined to express an initial position at conference (see Woodward & Armstrong, 1979; Epstein & Knight, 1998). This behavior led one critical justice to point out that, “all too damned often the Chief Justice will vote with the majority so as to assign the opinion, and then he ends up in dissent” (Schwartz, 1990, p. 14).
Although many claimed that Burger attempted to manipulate the Court’s agenda through opinion assignment, he was probably not the first chief to vote in this manner during conference. Indeed, Murphy (1964) traces sophisticated voting to control opinion writing to John Marshall. Although short on supporting evidence, some argue that Marshall wrote opinions “even in cases where he dissented” (Schwartz, 1993, p. 152). As Murphy (1964, pp. 84–85) hypothesized of chief justices when they were the first to speak and the last to vote, “Thus, before he finally commits himself, he knows where each Justice stands—at least for the present—and which side will most probably win. If his own views are going to be in the minority, he can vote with the majority and retain the opinion-assigning authority.”
Data drawn from the archives of Justice Lewis F. Powell (at Washington and Lee University) illustrate one particular strategy chiefs may use during conference—passing on their chance to cast the initial vote. Johnson et al. (2005) analyzed how often each justice passed on an initial vote in a sample of cases decided during the Burger Court. From their data it is clear that Burger, and then to a lesser extent, Rehnquist (when he became chief in 1986 ) passed on the initial vote significantly more often than associate justices. In fact, Burger was more than twice as likely to pass as any other justice on the bench. Rehnquist is also interesting as he passed 10 times more often once he was elevated to chief. Certainly the senior associate justices (Justice Douglas then Justice Brennan in this sample) passed more often than their colleagues (with the exception of Justice Stewart) but not nearly as often as chiefs.
The bottom line is that conference is a time when the justices finally speak to one another candidly about a case and cast their preliminary votes. While scholars have given this part of the Court’s decision-making process short shrift it is clearly important for how the justices decide. In fact, it sets the stage for the final part of the process—opinion writing.
After conference the writing process might seem straightforward, but this is not always the case. Keeping in mind that a majority opinion author must write an opinion that pleases at least four other members of the Court, the task is best described as a dynamic and malleable process whereby justices strategically maneuver in an attempt to satisfy their policy preferences (Maltzman, Spriggs, & Wahlbeck, 2000). In order to fully explicate the complexities of opinion writing, the process is described, the two main (and competing) models of opinion writing that define key actors who influence the final content of the majority opinion are summarized, and a case study of Gannett v. DePasquale—an exemplar of the opinion-writing and decision-making process—is presented.
Due to the volume of work and time constraints faced by the Court, many justices in the modern era use their law clerks to draft initial versions of opinions (Wahlbeck, Spriggs, & Sigelman, 2002; Peppers, 2006; Peppers & Zorn, 2008). Clerk’s legwork in hand, the opinion author prepares a first draft and, once complete, circulates this draft to the Court. At this point, no justice is bound by her stated policy preference or vote at conference; she instead has four options. She may join the opinion immediately, suggest some sort of change—via suggestion or threat, announce she is waiting before she takes any action, or circulate a concurrence or dissent (Maltzman, Spriggs, & Wahlbeck, 2000). The modal category is “join” (Maltzman, Spriggs, & Wahlbeck, 2000).
When a justice levels a threat or indicates a willingness to circulate a separate opinion the majority opinion author takes these indications particularly seriously. In particular, the existence of a well-crafted dissenting opinion has the potential (however small) to steal the majority, especially if it can satisfy the demands of the median justice (Lax & Cameron, 2007). Because justices wish to see their policies etched into law, majority opinion writers therefore seek to mitigate such potential threats. Warding off a dissenting opinion, however, can be difficult. Lax & Cameron (2007) employ game theoretic models to demonstrate the impact of time and effort costs on a would-be dissenter. In short, a majority opinion author must craft a sufficiently high-quality opinion that captures the median justice in order to prevent potential dissenting opinions from gaining traction.
No additional effort or response is necessary from the author when a justice immediately agrees to join his opinion. However, gaining the votes of justices who suggest changes to an opinion or who announce they are waiting, is a different matter. In these instances, scholars disagree as to which justices exert the most influence over the opinion author, and therefore over legal policy.
Scholars who investigate an author’s plight to command a majority often model opinion content as a corollary of ideology, pointing to two key players: the Court’s median member and the opinion author herself. The bench median model mirrors legislative models of the median-voter theorem in assuming policy converges on the median voter; it is, after all, this voter who must be captured in order to gain a majority and thereby create policy (Black, 1958). Applied to Supreme Court scholarship, the median justice is the swing vote and so may reject an opinion that falls too far from his ideal point. In fact, a majority opinion that does not perfectly satisfy the median justice allows the minority coalition an opportunity to craft such an opinion and capture this essential swing vote (Lax & Cameron, 2007). Extensive work by Spriggs and Hansford (2001) provides empirical support for this model. Their findings indicate that, indeed, the Court median exerts influence over the majority opinion.
Beyond the Court median, the opinion author exerts influence on the substantive outcome of a case. This agenda-control model stresses the importance of an opinion author in pulling opinion policy away from the median and closer to his own ideal point (Murphy, 1964; Rohde, 1972; Rohde & Spaeth, 1976; Slotnick, 1978; Slotnick, 1979; Brenner & Spaeth, 1988). Recent work by Bonneau, Hammond, Maltzman, and Wahlbeck(2007) adds to this body of work by incorporating a legal status quo. In this model, a justice’s decision to join an opinion is operationalized as a simple choice between the existing majority opinion and the lower court decision: any justice preferring policy offered in the majority opinion will sign on.
In sum, the rich body of literature on opinion writing provides ample evidence that the opinion author and the Court median exert particular influence over opinion content, and therefore policy. This is not to say the other seven justices are moot—to the contrary, the following example demonstrates the impact any single justice can have in this dynamic process.
Examining historical documents available for Gannett Co. v. DePasquale , 443 U.S. 368 ( 1979 )—a case involving freedom of the press and courtroom access—gives clear insight to this dynamic process. 29 To begin, the majority opinion was assigned to Justice Blackmun by Justice Brennan—this means that Justices Burger and Stewart were not in the majority when votes were counted at conference, but that both Brennan and Blackmun were. 30
On the 4th of April, Blackmun circulated a first majority opinion draft. The very next day Brennan signed the opinion after making a few minor suggestions, Stewart made known his intentions of writing a dissenting opinion, White declared he would await Stewart’s dissent before signing any opinion, and Stevens attempted to bargain with Blackmun. Specifically, Stevens believed Blackmun’s opinion went too far in favor of First Amendment rights (freedom of press), rather than Sixth Amendment rights (right to a speedy and public trial), and he was hoping Blackmun would back off this stance. In the meantime, Marshall joined Blackmun’s opinion and 13 days later Stevens joined Stewart’s dissenting opinion on the exact day it was circulated, noting, “I may add a paragraph of my own.” At this point, Blackmun was still two votes shy of commanding a majority.
Despite this deficit, Blackmun responded to Stewart’s dissent with, “The dissent merits a mild response. I shall circulate it later today.” Blackmun’s three-page response highlights the key differences in how each coalition wanted to set policy. Blackmun, along with Brennan and Marshall, wanted to set a clear legal ruling that carried a presumption of open access to trials as implied by the Sixth Amendment and considered preliminary hearings part of a trial. Stewart, and at this point, Stevens, felt the Sixth Amendment’s guarantee to a speedy and public trial was meant to protect the accused, not the public. Given this choice-set, and despite his earlier preference to await Stewart’s dissent, White joined Blackmun’s opinion creating a presumptive majority: Blackmun, Brennan, Marshall, White, and Powell. Although Powell had not yet declared any intentions, he had voted with this coalition at conference. This is not, however, how the majority opinion turned out.
On May 8th Chief Burger circulated a dissenting opinion, and also declared he would sign onto Stewart’s dissent with Rehnquist following shortly after. The following day, Powell sent a memo to Blackmun expressing doubts about how the majority opinion was taking shape. He too felt the crux of the issue was striking a correct balance between First and Sixth Amendment rights, but that Blackmun’s opinion tilted too far in favor of the press. Faced with these dichotomous policy choices, Powell eventually joined Stewart’s opinion and became the requisite fifth vote to transform that dissent into a majority opinion.
This example highlights two key aspects of the opinion-writing process. First, different justices employed different tactics in attempting to move policy closer to a preferred point. While White (the Court median) held out, some justices directly bargained Rehnquist and Brennan while others wrote separate opinions. Second, the central debate concerned specific aspects of policy, and not simply a decision to reverse or affirm the lower court’s ruling. This idea transcends the traditional liberal-conservative notion of decision-making to reveal a Court concerned with degrees and balancing rather than such simple dichotomizations.
The process by which the U.S. Supreme Court makes decisions has been the focus of a great deal of scholarly attention. Yet, there is more left to learn about our nation’s highest court. Scholars have only broken the surface of analysis about the Court’s most secretive meeting—the justices’ weekly conference. Further, there is growing debate about how the linguistic nature of justices’ words affects their decisions. While work has been done on how the language justices use during oral arguments impacts their decisions (Black et al., 2011; Johnson, Black, Goldman, & Treul, 2009) there is much work to be done. Indeed, scholars should focus on attorneys’ language during the arguments as well as in the briefs they submit to the Court (see, e.g., Corley, 2008). Further, scholars should consider how the choice of language in the Court’s opinions affects lower court decisions and how the public views the justices’ policy choices. In short, future work on the Court can and should continue to focus on the language of the law because it is the language that matters. In other words, how the Court rules (reverse or affirm) is simply not as important as the rules the justices set in their opinions. Linguistic analysis of our nation’s highest court is therefore the next frontier of judicial politics research.
U.S. Supreme Court justices are clearly political actors with policy preferences they seek to etch into federal law. However, they are limited in their ability to always reach their preferred outcomes by the fact that they do not make decisions in a vacuum and by the fact that rules and norms of behavior govern their decision-making process. While the Court is a unique institution, it is clear that it shares these characteristics with other courts within and beyond the United States.