Few substantive areas have merited as little empirical scrutiny as the Supreme Court's decisions on the conduct of United States foreign policy Many scholars have seemingly accepted as axiomatic that Court decisions on foreign policy have been rare and almost always supportive of the President. We challenge these twin assumptions and demonstrate that the Supreme Court has repeatedly issued decisions on the substance and process of American foreign policy, and that while generally supportive of the executive branch, the High Court has often ruled against it. Moreover, we model the outcomes of these decisions emphasizing the importance of the constitutional basis of laws and issues raised in judicial decision-making. We find that the executive branch is least likely to be supported when a case involved civil liberties, and when the executive powers claimed by the President were neither expressly permitted nor prohibited by the Constitution. The executive branch is more likely to emerge victorious when the case involved the supremacy of federal over state law, foreign actors, or the President's constitutional powers.
Few substantive areas have merited as little empirical scrutiny as the Supreme Court's decisions on the conduct of United States foreign policy The Supreme Court's edicts on domestic policy issues such as civil rights and liberties and economic regulations have received considerable scholarly analysis-, yet a systematic examination of this "high politics" domain has lagged far behind. The public law literature has provided us with extensive doctrinal analyses of Supreme Court decisions (Adler and George 1996; Franck 1992; Glennon 1990; Henkin 1990, 1996; Koh 1990), but many scholars have accepted as axiomatic that decisions with foreign policy implications are infrequent and that the Court "almost always" (Koh 1990) supports the Executive. We question these twin assumptions and demonstrate that the Supreme Court has often issued decisions where there are American foreign policy concerns in either the holding or the obiter dictum of the Court's reasoning, and that-while generally supportive of the executive branch-the High Court has often ruled against it. An analysis of the Court's foreign policy decisions is important not just because it investigates unexplored terrain or challenges conventional wisdom, it also enhances our understanding of Supreme Court decision-making.
The cases we analyze involve important questions regarding the President's Article 11 powers, such as treaty-making and war powers as well as the constitutional powers and rights of the coordinate branches of government and the citizenry. Whose privileges and powers triumph in these disputes? A majority of doctrinal analyses suggest Presidents prevail, but we demonstrate that there are issues on which the Supreme Court will support those opposing the Executive because of the types of constitutional claims advanced. Our analysis includes all cases with foreign policy implications since the Constitution's ratification, thus providing greater generalizability and historical insight into High Court decisions (Gates 1991; Epstein, Walker, and Dixon 1989). An exploration into the relationship between the Executive and the judiciary provides greater insight into the separation of powers and our system of checks and balances. Does competition among these institutions insure independence from one another, or is it possible that in some constitutional areas the Executive receives judicial deference?
We organize this article as follows. First, we review the literature on the Supreme Court and foreign policy to detail why the common wisdom encourages the stereotype that the justices show such deference to the executive branch. Second, based on doctrinal studies of judicial decision-making, we advance several hypotheses to predict when the Court supports the Executives position. Third, we discuss how we collected Supreme Court cases where foreign policy issues are raised and briefly explore trends in executive success. Our hypotheses are then tested on these data using probit analysis. Finally, we conclude with some thoughts regarding the applicability of our model of judicial support for the executive branch on other issues.
A HISTORY OF JUDICIAL DEFERENCE TO THE ExECUTIVE?
Many doctrinal, public law scholars have accepted the notion that the Courts have avoided cases where foreign affairs and policy are relevant (Adler and George 1996; Genovese 1980; Glennon 1990; Henkin 1990, 1996; Lindsay 1994). As Henkin (1996: 148) writes, "The Supreme Court in particular intervenes only infrequently and its foreign affairs cases are few and haphazard". David Adler (1996a) notes that when foreign policy issues do arise, the Court may recuse itself by claiming that the questions presented are non-justiciable, political questions (e.g., policy disputes between the President and Congress) or would involve the Court in matters beyond its competence, thus delegitimizing the Court and possibly tipping the balance of power inappropriately toward the judicial branch. In particular, we have seen the Supreme Court shy away from interfering with the President's war-making powers (Adler 1996c). Perhaps most importantly, some have argued that the Court wishes the nation to speak with one voice in foreign affairs and that a failure to do so would jeopardize national security (Adler 1996a; Henkin 1990, 1996). Such fears, it is contended, make the Court reluctant to challenge the executive branch.
Scholars have also advanced the claim that the Supreme Court finds in favor of the Executive when it does address foreign policy matters in its opinions. Adler argues that the Court has aided the President by inferring legislative support for his actions from congressional inaction in certain types of issue areas. Where express constitutional or statutory authority has not existed, the Court has assumed that a lack of opposition constitutes implied acquiescence and has supported Executive action. Thus, Koh (1990) and others have argued that the President "almost always" wins in foreign affairs when such cases do arise.
In addition, presidential scholars typically concur with one or both of the claims of judicial abstention and presidential supremacy (Corwin 1957; Cronin 1975; Schlesinger 1973). Some, like Rossiter (1956: 38), argue that "...the President may act as if the Supreme Court did not exist" and minimize the judicial branch's role in foreign policy matters (see also Corwin 1957; Hamilton 1845; Schlesinger 1973). Others acknowledge a degree of Supreme Court influence, especially in landmark cases (e.g., United States v. Curtiss-Wright Export Corp [1936. 299 U.S. 304] and Goldwater v. Carter [1979. 444 U.S. 9961), but view the Court as ". . . aiding and abetting an expansive view of presidential power" (Cronin and Genovese 1998: 225). Research on presidential decision-making in foreign policy accepts as axiomatic presidential dominance over the Congress and treats the judicial branch as peripheral at best (Crabb and Holt 1988; Hinckley 1994; McCormick 1992; Petersen 1994).
Empirical research in public law and judicial politics has also marginalized the role of the Courts in foreign policy matters. As Ducat and Dudley (1989: 99) note ". . . there remains a virtual absence of any empirical study of federal judicial decision-making on the issues of presidential power as distinguished from research about federal judicial support for policy items that comprise the President's agenda".
Should scholars accept these assertions of judicial abstention and presidential dominance in foreign policy matters? In fact, there are some doctrinal scholars who argue that the High Court's decisions are fashioned to serve a broader view of institutional purposes and thus obscure the foreign policy implications of many cases. The justices have several pragmatic ways to circumvent admitting that a foreign policy issue before them might involve political questions they have been reluctant to consider. They may do this by "reclassifying" a case as something other than foreign affairs or by simply paying lip service to the political question doctrine while proceeding to decide a case on the merits (Franck 1992). Scholars also note that while the political question doctrine is often used to justify judicial abstention, ". . . there is little agreement as to anything else about it-its constitutional basis-, whether abstention is required or optional-, how courts decide whether a question is 'political' and which questions are" (Henkin 1996: 144). Such ambiguities may give the Court the latitude it requires to decide a foreign policy issue on its merits, even as the justices proclaim their reluctance to address the real issue.
In fact, there are several reasons for undertaking a comprehensive, empirical analysis of the Supreme Court's doctrine in this issue area. In the absence of any systematic search in legal databases for evidence of judicial decision-making on foreign policy, we cannot conclude the Court is reluctant to rule on these matters. We know that foreign policy issues encompass more than those issues mentioned in the enumerated powers of the Executive, which have traditionally delimited the universe of cases many scholars examine. Foreign policy matters also involve international trade, immigration, Acts of State, and travel, to name but a few. Should we also assume Presidents are as likely to win in cases with lower issue salience such as these? In fact, we argue that by distinguishing among these types of issues that implicate foreign policy, we can determine where the executive branch's claims are strongest, and on what issues the claims of other political actors may carry greater weight before the High Court.
A MODEL OF SUPREME COURT ADjUDICATION IN FOREIGN POLICY
Scholars have long recognized the importance of the issues raised in disputes in influencing judicial outcomes, but behavioral research has only recently begun to incorporate these issues into decision-making models (Brace and Hall 1990; Epstein, Walker and Dixon 1989@ Gates 1991; Hall and Brace 1989, 1992; Smith 1988). According to Ducat and Dudley (1989: 102), "Whether we are talking about spheres of action where presidential decision-making is constrained by rules or areas where judges exhibit virtually categorical deference to his discretion, it is clear that the identity of the policy area will be determinative" (emphasis added).
In particular, Ducat and Dudley (1989) examined federal district court decisions involving presidential power from 1949 to 1984, and found that the distinction between foreign policy and domestic policy issues is critical in determining outcomes (see also Genovese 1980). Courts are conscientious of the President's constitutional powers in foreign policy and so judges are more likely to defer to his command authority (Genovese 1980; Yates and Whitford 1998). Conversely when issues arise regarding domestic policy, the courts appear to be swayed by political factors such as presidential popularity and their own partisan composition. This distinction between foreign and domestic policy deference by the Courts is also mirrored in the "two presidencies" thesis which supposes that Presidents find higher levels of congressional support on foreign policy rollcall votes (Wildavsky 1966).
If the distinction between foreign and domestic policy issues is important in determining judicial support for the executive branch, we should search within foreign policy cases to determine where executive claims are most likely to prevail. While there is nothing that may be properly termed a theory of judicial decision-making in the doctrinal analyses of foreign policy decisions, there is a fair degree of consensus regarding the Court's propensity to side in favor of or rule against the President on specific types of foreign policy issues.
Many scholars note the primacy and weight justices attach to constitutionally based claims of authority While the enumerated powers of Article 11, section 2, provide the President with his strongest claims in foreign policy cases, the Supreme Court has also consistently attempted to demonstrate that the Constitution should be read so as to provide the national government with great discretion in foreign affairs, especially vis-a-vis the states (Henkin 1990, 1996; Silverstein 1997). When executive branch claims are not based on either the enumerated powers or the vertical separation of powers, but are based instead on a hypothesized grant of authority (e.g., where the Constitution and statutory law are silent), there is apt to be greater disagreement among the justices regarding the proper scope of executive authority (Silverstein 1997). When other political actors, such as the Congress or individuals advance constitutionally based claims for powers and rights, the Supreme Court must weigh the protections afforded by the Constitution to these actors against the needs of national security. Thus, we should not expect the Executive consistently to prevail in such disputes.
Before proceeding, however, it is useful to describe what our model is not. First, while scholars recently have begun to make use of "political" factors in models of judicial support for the executive branch (Ducat and Dudley 1989; Yates and Whitford 1998), such as presidential popularity and partisan ties between Presidents and justices, these scholars also argue that such factors play little if any role in cases involving foreign policy (in fact the relationship between popularity and presidential victories is negative in foreign policy cases). Second, we do not apply the attitudinal model (Segal and Spaeth 1993) to studying foreign policy cases. Since foreign policy involves many types of issues, such as federal powers, civil liberties, and commerce, it is difficult to speak of foreign policy cases along one ideological continuum. Researchers have also studied support for the Solicitor General in the Supreme Court across a variety of issues (Segal 1988, 1990; Segal and Reedy 1988); however, most of this research focuses on the votes of individual justices and the role of ideology. We wish to focus on the Court's role in providing institutional outcomes and the importance of issues. Lastly, our model is not a "legal" model. We do not rely on "plain meaning," intent, precedent, and balancing factors associated with the legal model. Instead, we look to the issues raised in foreign policy opinions as predictors of the strength of the Executives constitutional and statutory arguments. We discuss in the conclusion how our model relates to these other perspectives.
PREDICTING SUPREME COURT DECISIONS INVOLVING FOREIGN POLICY
Constiutional Claims of Authority
Enumerated Executive Powers: The Constitution provides the President with three specific grants of authority in the realm of foreign policy First, he is designated Commander-in-Chief of the armed forces and the state militias when called into service. Although Congress is given the power to declare war and provide for the armed forces, the President is entrusted with waging war. As some have argued (Adler 1996b; Fisher 1995), the framers intended this power to be narrowly construed; and throughout much of early American history the courts typically adhered to this strict interpretation. In recent years, however, the Supreme Court has more often than not sided with the President either on the merits or by dismissing cases on the basis of non-justiciability, thereby preserving the President's power (Crabb and Holt 1988). Second, the President under Article 11, section 2, shares-with the Senate's advice and consent-the power to negotiate and make treaties. Again, scholars find that the courts have generally sided in favor of the President in such cases and others involving executive agreements (Silverstein 1997). Third, the President has the power to receive foreign ambassadors. Although perhaps designed to be a purely ceremonial function (Adler 1996c), this duty has evolved into the more powerful ability to recognize foreign governments. Therefore, we argue that when one of the three-war power/Commander-in-Chief, treaty power, and recognition power-arises in a dispute, the Supreme Court will be more likely to support the President. We employ three separate dummy variables for each constitutional power.
Coding procedures for these and all other variables are discussed in the Appendix. it should be noted, however, that we examined both the Supreme Court opinion and lower court opinions separately to determine the issues and actors involved in a dispute in order to provide greater reliability for our independent variables.
Federal Supremacy: Deeply embedded within our form of government is the primacy of federal institutional power over the authority of state governments. Almost all scholars agree that when the executive branch and lower-level governments are on opposing sides, or the issue concerns the supremacy of federal law or international treaty over state laws, the Executive typically prevails (Henkin 1990, 1996:150). According to Henkin, "The language, the spirit, and the history of the Constitution deny the States authority to participate in foreign affairs, and its construction by the courts has steadily reduced the ways in which the States can affect American foreign relations." Recently this has been given emphasis by Silverstein (1997) who argues that the Court has grounded executive branch authority in foreign policy in part on the vertical separation of powers--the relationship between the national and the state governments. The Court has sought to justify deep and wide-ranging powers for the Executive so that it may act with great discretion in foreign affairs against the multifarious interests of the states. Therefore, we hypothesize that in cases involving states' rights, the executive branch will generally win. All cases where states were parties or where the opinion explicitly raised the issue of states' rights relative to national authority were coded 1, other cases were coded 0.
Foreign Actors: Under section 2 of Article III of the Constitution, not only may domestic parties rely on "judicial power," but "foreign states, Citizens, or Subjects" may also seek the protective veil of the judicial branch. Does the Constitution, however, provide such actors with the same privileges and protections as domestic interests? In fact, public outrage over the Courts decision in Chisholm v. Georgia (1793. 2 Dallas 419) to protect a British creditor over the state of Georgia, prompted the passage of the Eleventh Amendment (Abraham 1993). More importantly, given that U.S. relations with foreign actors are governed by international treaties, and given the weight the Court attaches to the Executives interpretation of treaties (e.g., Hamilton 1845), it is only reasonable to suppose that the Court will tend to find in favor of the Executive over the claims of foreign actors, whether public or private (see also Franck 1992). In cases involving foreign governments, businesses, and individuals, we hypothesize that the Court will more likely support the executive branch. The foreign actor variable is coded I when a foreign actor opposed the executive branch; otherwise it is coded 0.
Other highly salient legal issues, however, may not involve express constitutional powers that the Court may view as settled, or the bench may avoid via the political question or standing doctrines. Rather, powers that are neither expressly permitted nor denied to the executive branch lie in a gray zone of executive authority In cases where the basis for the Executive's prerogative power is nebulous, the Court possesses greater latitude to circumscribe its authority and may use the opinion to clarify explicitly that the Executive does not possess the prerogative power under the circumstances presented. Thus, we take a position that is consistent with Silverstein (1997: 176) and argue that it is precisely these cases where the Court possesses the opportunity to circumvent the obstacles of the "political question" doctrine and non-justiciability to reach a decision on the merits. We looked to court opinions (whether in the holding, reasoning, or the obiter dictum) to tell us when the issue at stake involved questions of executive authority that were not specifically addressed by the Constitution or statutory law. We hypothesize that when the majority opinion specifically stated that the Constitution and statutory law were silent on the type of executive power at issue, the President will be more likely to lose.
Contested Constitutional Claims
Civil Liberties: Civil liberties may be the most formidable, constitutional issue where the executive branch has confronted counter-claims of rights and protections, because the Bill of Rights was specifically enacted to limit federal intrusion and power. When the Court has ruled against the executive branch in the post-1937 era, it has often done so because it claims executive actions violated fundamental rights that are guaranteed under the Constitution, especially those of freedom of speech and press (Franck 1992; Glennon 1990; Henkin 1996). While this emphasis on civil liberties is a relatively new phenomenon in the Court's historical development, it is consistent with the transformation of the Supreme Court's ability to assert its own institutional authority, and its role as guardian of individual rights (Elliott 1974). Therefore, we hypothesize that whenever an opinion raises the issue of civil liberties, the executive branch will be more likely to lose. All cases involving claims arising under the First, Fourth, Fifth, Ninth, or the Fourteenth Amendments to the Constitution are considered to be cases involving civil liberties and are coded 1. All others are coded 0.
The Separation of Powers and the Congress: Although some (e.g., Glennon 1990) argue that Congress always prevails over the President in foreign policy disputes that reach the courts, the High Court has more often than not been the President's ally. It has acquiesced to broad legislative delegation of foreign policy power to the executive branch. It has allowed Presidents to act in the absence of congressional action. It has increasingly found that where the Congress does not explicitly forbid presidential authority, the President is assumed to have discretion (Adler 1996; Silverstein 1997). Therefore, we hypothesize that when the Court explicitly notes that the issue involved the separation of powers between the Executive and the legislature, it will be more likely to support the Executive. The variable is coded I if the opinion raises the separation of powers issue. Otherwise it is coded 0.
To begin this examination, we conducted a search of the WESTLAW database containing all Supreme Court opinions from 1790 to 1996 @ We searched for specific references within the Court's opinions to foreign policy issues and powers. Thus, we let the Court's own language indicate whether or not foreign policy concerns are raised. If the Court is reluctant to address foreign policy issues, such a strategy may result in an under-counting of the number of foreign policy cases. It is, however, a conservative strategy because we do not rely on the opinions of others (e.g., the Solicitor General, or amicus curiae), who may have their own reasons to elevate the significance of a case's foreign policy ramifications. Therefore, this approach provides a more rigorous test of our contention that foreign policy cases are not rare. The key words we used in this search were: foreign policy, foreign relations, foreign affairs, international policy, international affairs, international relations, national security, national defense, commander-inchief, and international law We examined both plenary opinions and summarily decided cases where the Court discussed the substantive dispute.
The initial search resulted in approximately 750 potential matches. We examined each case to determine how the foreign policy reference was addressed in the majority opinion.1 We excluded all cases where the foreign policy concern was indirect because it was simply a reference to a name, title, or subject of a congressional committee, public law, or prior decision. Similarly, we excluded a case if the reference was remote because the foreign policy referent was contrasted with domestic policy issues that were squarely presented within the case (e.g., Baker v. Carr [1961. 369 U.S. 1861; Marbury v. Madison [1803. 5 U.S. 1371).
Because we were concerned only with those cases where executive branch powers were directly implicated by the decision, we include only cases where the U.S. government was a participant or the Justice Department filed a brief that clearly demonstrated the party that the federal government supported. Ultimately, we selected only those cases where in the majority opinion the Court raised a foreign or defense policy issue involving the powers of the executive branch and a representative of the federal government either argued the case or submitted an amicus curiae indicating which party the executive branch supported. Despite these stringent criteria, we found 347 foreign policy cases decided between 1790 and 1996. All case citations are available upon request. A test of intercoder reliability regarding case inclusion showed 93.3 percent for the two coders. To provide even greater validity and generalizability to our analysis, we analyze the results from a much more restrictive search involving presidential powers later in the article.
Some may wonder why we did not rely on the issue codings contained in the U.S. Supreme Court judicial Database. First, the Spaeth Database does not contain an issue category for foreign policy cases, just as, for example, it does not contain one for environmental or agricultural policy. Rather, cases are coded according to the legal issues involved, which may be somewhat different than the broader policy issues addressed. Second, the Spaeth Database only extends back to the mid- 1940s. We are interested in studying judicial decision-making on foreign policy cases since the U.S. Constitution was ratified. Third, we were interested in cases where there are foreign policy ramifications rather than only those cases where foreign policy is the direct issue before the Court. Our approach is consistent with doctrinal analyses of Supreme Court cases because future decisions may rely upon the full reasoning of the Court (including obiter dictum) to justify either limiting or expanding presidential power (Silverstein 1996; Horwitz 1987). This necessitated developing our own search criteria.
We often find scholars arguing that there are relatively few cases involving executive power in foreign policy, and that only on limited occasions will the High Court interfere with the executive branch's preferences. Quite simply, we do not find this to be the case. Relying upon the Court's doctrinal statements, we find many cases (347) throughout two centuries of Supreme Court jurisprudence. This would not seem to support the notion of a reluctant Court. The conventional wisdom that the executive branch customarily prevails in these cases can also be questioned. The executive branch does typically prevail in these cases-it is victorious 69.1 percent of the time-but perhaps not to the degree that one might otherwise believe from doctrinal analyses of executive power. While we have no direct, quantitative basis of comparison between others' anecdotal claims and our numbers, these findings challenge those who argue that, "The president almost always seems to win in foreign affairs" (Koh 1990: 148). This percentage is also lower than what Segal (1988) finds in his examination of Solicitor General victories before the Court. His results indicate that the Solicitor General won 75.3 percent of the time on amicus briefs between 1953 and 1982. Yates and Whitford (1998), however, find that Presidents win 62.2 Percent of the time on foreign policy cases from 1949-1993-slightly less than our percentage.
The Figure displays the percentage of executive branch victories over time aggregated by decade. We should note at the outset that the infrequency of cases before the Civil War causes the percentages to change rather dramatically. In general we see that while the percentage of victories fluctuates over time, it hovers in the range of 60-80 percent. In recent years the executive branch lost more frequently in the 1930s, 1950s, and 1960s, while it has been winning at historically high levels of late. It may well be that this unprecedented record of recent victories gave rise to the contemporary view that Presidents are nearly invincible on foreign policy issues before the Court. Certainly in direct contrast to the Congress, which took steps to restrain the President (e.g., the Case Act and the War Powers Resolution), and the public, which grew increasingly distrustful of executive power after Vietnam and Watergate, the Court appears to have aligned itself more closely with the President (Crabb and Holt 1988).
We also examined the differences in executive outcomes before and after two watershed events in the history of executive power, the Curtiss-Wright decision (1936), and the onset of the Cold War, defined here as beginning in 1948 and ending in December of 1991 with the break up of the Soviet Union (results not shown). Before Curtiss-Wright, the Supreme Court supported the Executive 68.8 percent of the time. After the decision, this increases to just 69.4 percent. So while the Curtiss-Writht case may have been viewed by doctrinal scholars as signaling a vast expansion of presidential prerogative, the level of judicial deference to the executive branch does not appear to have changed to any great extent (Silverstein 1997). Finally, and quite interestingly, before the Cold War the executive branch was supported 69 percent of the time, while after-ward, the percentage of victories increases to only 69.3 percent. While this difference is too slight to justify changing our views about the Cold War's impact on presidential power, at a minimum it suggests that the Court did not provide the Executive with blank checks to confront the communist threat.
Our unit of analysis is the Supreme Court decision. Specifically, we ask whether the Supreme Court favors the executive branch on foreign policy cases, or the arguments of the opponents of executive power. We include all foreign policy cases that involved the executive branch and were argued by the Solicitor General or a representative of the executive branch (e.g., lawyers for the Department of Defense, etc.). Table I contains the entire universe of Supreme Court decisions with foreign policy implications and employs variables that were coded from the Supreme Court opinions.
Table 2 contains only those cases for which there was an available prior or lower court decision. We use those opinions to code the independent variables (the dependent variable in each table is still the decision in the Supreme Court).2 Because it is possible that the Supreme Court's decision outcome may influence the types of issues discussed in its opinion, we include information from the lower courts in order to provide greater reliability in the coding of our independent variables. Since lower or prior court records were not available for all the Supreme Court cases, the N in this table is smaller. We found that our results did not vary significantly depending on whether or not the executive branch was a direct participant in the case, was arguing the case for another party on its own initiative, or was filing an amicus brief in favor of a litigant.' Because the dependent variable is dichotomous, we utilize probit analysis.
Both the overall fit of the model and the statistical significance of most of the coefficients demonstrate that it is possible to predict the outcomes of Supreme Court as an institution in foreign policy cases. The model predicting outcomes for the data based on Supreme Court opinions is correct 74.6 percent of the time for an approximately 17.7 percent reduction in error over predicting the modal category every time (support for the executive branch, which occurs 69.1 percent of the time). As shown in Table 2, the percentage of correctly predicted cases using data based on lower court opinions is 74.6 percent, which represents a 20.1 percent reduction of error. In addition, the likelihood that all the coefficients are equal to zero is miniscule, according to the log likelihood ratios.
We readily see that the executive branch is significantly more likely to be supported in cases involving constitutional powers. Both the variables measuring war powers and treaty powers are statistically significant and positively signed in Table 1, although only the war powers variable is significant in cases using lower court opinions for issue coding. As we expected, the justices endorse presidential preferences more often when the issue pertains to the enumerated powers of Article 11. We would expect that given the increasing amount of discretion given to the Executive to use force, both by the courts and the Congress, that such issues would most favor the chief foreign policymaker. The only discrepancy we find between the results obtained from using the Supreme Court versus lower court opinions to code the cases is the treaty variable. Its statistical insignificance in the second set of estimates indicates that the Supreme Court is more likely to highlight such powers when supporting the Executive, while such issues tend not be raised in lower court opinions even if the President is being supported. Still, we find that Presidents win 82.4 percent of the time treaty powers are raised in Supreme Court opinions, and 78.6 percent of the time when they are raised in tower court opinions-a fairly small difference.
Generally speaking, the Court has sided in favor of the national government when state laws and actions conflict with federal law. This variable is statistically significant when using both the Supreme Court and lower court opinions to code the cases. State laws or actions that impair the furthering of national policy ends are likely to be overturned or interpreted in such a way as to preserve national dominance. When holding all other variables constant at their mean values, the probability of an executive branch victory in such cases increases by 18 percent in Table I and 19.3 percent in Table 2. This finding further illustrates the importance of the Court's role in the vertical separation of powers (Silverstein 1997). In addition to its role as protector of the balance of power among the three national branches of government, one of the most important functions of the Court has been to assure consistency of law between federal and state governments, especially as it relates to foreign policy issues. Only by such action can the national government speak with one voice to foreign governments and other international political actors.
We also see that the executive branch has been particularly successful in disputes involving citizens and corporations of foreign states as well as foreign governments. Not surprisingly, the Supreme Court appears to accord greater deference to the claims of American interests than it does to foreign interests. The foreign actor variable is statistically significant in both Tables and increases the likelihood of executive support by 16.6 percent in cases coded using the Supreme Court opinion and 25.6 percent in cases using the lower court opinion, holding other variables constant at their mean value.'
When executive powers conflict with civil liberties, the Supreme Court tends to take the side of individual rights. This finding supports our conception of the Court's discretion to rule against the executive branch in cases where there are contested rights and authority In cases involving the right to travel, communist subversives at home, and the publication of information related to national security, the Court has usually sided in favor of the Bill of Rights. When cases involve such rights, the probability of an executive branch victory declines by 20.6 percent in cases coded with the Supreme Court opinion and 22.5 percent in cases coded using a lower court opinion. Especially noteworthy in this regard were the decisions handed down in the late 1950s and the 1960s when many of the tools given to the President to deal with communists in the government and sensitive industries and to deport foreign subversives were battered in a series of far-reaching Court decisions.
Where presidential authority is in doubt, the Court is more likely to side against the White House. When the Court finds that a particular executive policy is neither expressly permitted nor prohibited by the Constitution, it often finds against the President. This variable is statistically significant in those cases coded using the Supreme Court opinion, but statistically insignificant in the other estimates. When a case involves ambiguous authority it decreases the likelihood of presidential support by 23. 1 percent in the first set of estimates. When the Court makes such a determination, it does not appear to be acting as the champion of presidential authority that many scholars have discussed. Instead, the Supreme Court has often struck down presidential actions that have pushed the envelope of legitimate authority too far, as was evidenced in the famous steel seizure case.
We found evidence in 17 Supreme Court opinions of such reasoning, and found Presidents lost nine of these cases. Given that this variable is statistically insignificant in the estimates that make use of lower court opinions to code the cases, it is possible that the Supreme Court is merely using rhetoric about the nebulous nature of presidential power in some cases to rationalize its opinions. But it may also be the lower courts that, in dealing more with case facts than theories regarding the nature of presidential power, avoid such constitutional disputes. They may view such issues as best left to the Supreme Court to resolve.
The Court does not appear to rule against the executive branch when an issue concerns the Congress and the separation of powers. This variable is statistically insignificant in both sets of estimates, indicating that discussion of the possibility of a violation of the horizontal separation of powers does not mean the President's powers are in need of protection by the Supreme Court. Because our measure does not allow us to determine whether or not members of Congress themselves believed legislative powers were at stake in a decision, we also examined the impact of congressional participation in a case (as parties or in amici briefs) by either the majority or minority leaderships of the House and/or the Senate (results not reported). Such involvement has been a fairly recent phenomenon as Congress's ability and will to challenge the Executive has increased with additional resources and members' displeasure over Presidents' disregard of Congress's foreign policy powers. Again, we found no statistical evidence of judicial deference to members' interests. Since there were relatively few such cases (six), we cannot infer that the Supreme Court is more the ally of the executive than the legislative branch.
We are also interested, however, in those cases where the specific powers of the Presidency, not just the executive branch, were addressed in the Courts majority opinion.' That is, we wished to distinguish between cases where the Court addressed the authority of the whole of the executive branch (which constitute our universe of all foreign policy cases), and those where the powers of the President specifically were at issue. While we still believe the first set of cases has consequences for executive authority because the Solicitor General or other executive branch lawyers argued the case, a more narrowly focused search should provide a more demanding test of the primary arguments of this article. Therefore, we examined each case to determine if, (1) the powers of the President or Presidency were specifically referred to in the headnotes appearing at the beginning of each opinion, or (2) the powers of the President were afforded discussion (at least one paragraph) in the majority opinion (Yates and Whitford 1998). Using these criteria, we winnowed our data set down to 107 cases in the 1790-1996 period. We reran the model, albeit with one minor adjustment. We found that Presidents won every case involving treaty powers; consequently we were forced to drop this variable. The results are shown in Table 3. We elected to use the issue codings based on the Supreme Court opinion, since the results obtained from using lower court opinions were substantially the same.
We find that many, although not all, of the relationships posited above continue to hold in this probit model. We see that Presidents win slightly fewer of these cases (65.4 percent) than they do in the full model (69.1 percent). The model correctly predicts 72 percent of the cases for a reduction in error of 21.3 percent. Again, when a case involves war powers, Presidents are more likely to win, although they are not more likely to emerge victorious when the dispute involves the recognition power. Presidents are not statistically more likely to win when they are opposed by foreign entities or by state governments. This discrepancy with the results obtained in the prior estimates may be due to the lack of cases involving state and foreign actors, for we still find the percentages of presidential victories vary little between the two types of cases for both variables. Finally, Presidents are less likely to win when a case involves civil liberties and when they are neither expressly permitted nor denied some power by the Constitution.
Thus, not only do we find support for our hypotheses in the universe of all foreign policy cases from the founding of the Constitution to the present day using both the Supreme Court and lower court opinions to code our variables, our model also holds up when we restrict ourselves to disputes directly implicating the powers of the President himself. That such results hold up across 200 years of Supreme Court cases seems to challenge the conventional wisdom regarding the Court's reluctance to address cases where there are implications for foreign policy, and its unwillingness to rule against the President.
Our objectives in this article have been threefold. First, we sought to demonstrate that the Supreme Court has often decided cases involving foreign policy issues throughout its history. Previously, most scholars assumed such rulings were rare because of the political question doctrine, non-justiciability, and the justices' perception that such affairs were beyond their competence. We found evidence of 347 cases with implications for U.S. foreign policy and the executive branch from 1789 through 1996 that we believe demonstrates that the Court does not avoid such issues. Second, we sought to show that the Supreme Court has not always been the steadfast ally of the executive branch. Rather, we demonstrated that the Court has found against the position of the executive branch in 107 of our 347 cases. While the executive branch's record is impressive, it does not "almost always win." Finally, we sought to demonstrate that it is possible to develop a model of decisional outcomes within the foreign policy domain.
Employing an issue-based model of judicial outcomes, we found that the executive branch was least likely to be supported when a case involved civil liberties and when the executive powers claimed by the President were neither expressly permitted nor prohibited by the Constitution. The executive branch was more likely to emerge victorious when the case involved the President's constitutional powers, the supremacy of federal over state law, and when the case involved foreign actors. Based on these findings we believe that further investigation of Supreme Court adjudication of foreign policy would yield additional insights into judicial behavior. Future research should explore the reasons why lower courts are more apt to find against the Executive, why the Court's docket and support for the Executive have greatly increased in recent years, and how U.S. foreign policy history has affected the Court.
More importantly for the community of judicial scholars is the generalizability of our model. We believe it is useful to conceive of court cases is pertaining to specific or general types of policies, not just areas of the law If we believe that justices have policy preferences in the same way Presidents and members of Congress do, it only makes sense to examine the Court's caseload as comprising a policy agenda as well as a legal docket. We have illustrated that by examining a set of diverse legal cases that share a common, issue property-their relevance to foreign policy-we can uncover patterns in Court outcomes that might go unnoticed if we had examined such cases according to the legal issue presented. This view certainly comports well with those who view the Court as a political institution and the justices as policy-driven decision-makers.
Thus, we believe this approach is not at odds with other models of judicial behavior, particularly the attitudinal model. Our approach is intended to supplement rather than supplant. Even though we do not test for the impact of judicial attitudes on foreign policy outcomes, it is possible to speculate about when ideology is relevant. For example, the justices' attitudes may matter most when a foreign policy issue does not concern one of the enumerated, "rule-based" powers of the executive branch. justices may perceive greater latitude to employ their own values in such cases without fear of precipitating a constitutional crisis. We do not deny that attitudes or politics matter-, we do suggest that the impact of such factors may be issue dependent as others have shown (Ducat and Dudley 1989- Yates and Whitford 1998). We also show that litigant status (e.g., states and foreign actors) is an important determinant of judicial outcomes as many others have demonstrated (Caldeira and Wright 1988; Epstein 1985). Finally, we believe that the general insights derived from our model can be used to predict judicial outcomes in a wider variety of cases. For example, by distinguishing among issues that involve enumerated powers and rights, and those involving ambiguous and contested powers, scholars should be able to determine when executive branch claims of authority are most likely to prevail.
1 Again, we wanted to be quite conservative and thus were interested only in majority reasoning because it provides the greatest precedential Support for future cases that may also have ramifications for foreign affairs. Thus, we did not use a case if the language was found in either the concurring or dissenting opinions. Only when there was a per curiam opinion did we examine concurring opinions.
2 In addition, we selected from our original universe of cases only those where the lower court record specifically mentions at least one of our search criteria (results not shown). We examined every lower court decision-"same case" citations for a discussion of one of the key terms ("connected case" citations were excluded). We selected a case if either the district or appellate court record discussed one of our referent terms used in the original selection. This search captured most, but not all of the cases we used in Table 2 (in some of those cases a lower court record did exist, but the opinion did not mention one of our search criteria). This set of criteria for selecting cases provides a set of 114 cases and confirms most of our previous findings (results not shown). The only major deviation was the change in sign on the state variable. In this set of estimates it becomes negative, although when we examine a breakdown of such cases, Presidents still won six out of the eleven cases involving states.
3 We also examined Congressional Quarterly's Guide to the Supreme Court for its listing of important cases to determine how well Presidents fared. In fact, we found that Presidents did worse on this subset of our cases (46 cases or 13.2 percent of our cases). They won 54 percent of these important cases, while they won 69.1 percent of our foreign policy cases. Thus, it would be difficult to make the argument that the Court is more likely to find in favor of Presidents in such crucial decisions in order to avoid handing Presidents damaging losses.
4 Some have wondered when the Court would ever find in favor of a foreign actor. Those decisions which hand the Executive a loss and a foreign entity a win cover a variety of disputes involving different foreign entities. Cases range across all time periods of American history and include foreign individuals, foreign sovereigns, as well as cases involving U.S. intervention for both domestic and foreign parties.
5 This approach is similar to Yates and Whitford (1998: 540) who examined the Spaeth data set, and then selected cases based on whether presidential power received "substantial discussion" ("typically a paragraph or more" and where the discussion was "integral to the case's disposition".
Abraham, Henry J. 1993. The Judicial Process: An Introductory Analysis of the Courts of the United States, England, and France, 6th ed. New York: Oxford University Press.
Adler, David G. 1996a. "Court, Constitution and Foreign Affairs." In David G. Adler and Larry N. George, eds., The Constitution and the Conduct of American Foreign Policy. Lawrence: University of Kansas Press.
. 1996b. "The Constitution and Presidential War Making." In David G. Adler and Larry N. George, eds., The Constitution and the Conduct of American Foreign Policy. Lawrence: University of Kansas Press.
. 1996c. "The President's Recognition Power." In David G. Adler and Larry N. George, eds., The Constitution and the Conduct of American Foreign Policy. Lawrence: University of Kansas Press.
Adler, David G., and Larry N. George. 1996. The Constitution and the Conduct of American Foreign Policy. Lawrence: University of Kansas Press.
Brace, Paul, and Melinda Gann Hall. 1990. "Neo-Institutionalism and Dissent in State Supreme Courts." Journal of Politics 52: 54-70.
Caldeira, Gregory A., and John R. Wright. 1988. "Organized Interests and Agenda Setting in the U.S. Supreme Court." American Political Science Review 82: 1109-27.
Corwin, Edward S. 1957. The President, 4th ed. New York: New York University Press.
Crabb, Cecil V, Jr., and Pat M. Holt. 1988. Invitation to Struggle: Congress, the President and Foreign Policy, 3rd ed. Washington, DC: Congressional Quarterly Press.
Cronin, Thomas Ed. 1975. The State of the Presidency. Boston: Little, Brown. Cronin, Thomas Ed, and Michael A. Genovese. 1998. The Paradoxes of the American Presidency. Oxford: Oxford University Press.
Ducat, Craig R. 1996. Constitutional Interpretation, 6th ed, Minneapolis: West Publishing.
Ducat, Craig R., and Robert L. Dudley. 1989. "Federal District Judges and Presidential Power During the Postwar Era." journal of Politics 51: 98-118.
Elliott, Ward E. 1974. The Rise of Guardian Democracy: The Supreme Court's Role in Voting Rights Disputes, Cambridge, MA: Harvard University Press.
Epstein, Lee. 1985. Conservatives in Court. Knoxville: University of Tennessee Press. Epstein, Lee, Thomas G. Walker, and William J. Dixon. 1989. "The Supreme Court and Criminal Justice Disputes: A Neo-Institutional Perspective." American Journal of Political Science 33: 825-41.
Fisher, Louis. 1995. Presidential War Power. Lawrence: The University of Kansas Press.
Franck, Thomas M. 1992. Political Questions/judicial Answers. Princeton: Princeton University Press.
Gates, John B. 1991. "Theory, Methods and the New Institutionalism in judicial Research." In, John B. Gates and Charles A. Johnson, eds., The American Courts: A Critical Assessment. Washington DC: Congressional Quarterly Press.
Genovese, Michael. 1980. The Supreme Court, The Constitution and Presidential Power Lanham, MD: University Press of America.
Glennon, Michael J. 1990. Constitutional Diplomacy. Princeton: Princeton University Press.
Hall, Melinda Gann, and Paul Brace. 1989. "Order in the Courts: A Neo-Institutional Approach to Judicial Consensus." Western Political Quarterly 42: 391-407.
-. 1992. "Toward An Integrated Model of Judicial Voting Behavior." American Politics Quarterly 20: 147-68.
Hamilton, Alexander. 1845. The Letters of Pacificus and Helvidius with the Letters Of Americanus. A Facsimile Reproduction. Delmar, NY: Scholars' Facsimiles and Reproductions.
Henkin, Louis. 1990. Constitutionalism, Democracy and Foreign Affairs. New York: Columbia University Press.
. 1996. Foreign Affairs and the Constitution, 2 ed. Oxford: Clarendon Press.
Hinckley, Barbara. 1994. Less than Meets the Eye. Chicago: University of Chicago Press.
Horwitz, Morton J. 1987. Transformation of American Law, 1870-1960: The Crisis of Legal Orthodoxy, New York: Oxford University Press.
Koh, Harold Hongju. 1990. The National Security Constitution. New Haven: Yale University Press.
Lindsay, James. 1994. Congress and the Politics of U.S. Foreign Policy. Baltimore: Johns Hopkins University Press.
March, James G., and Johan P Olsen. 1984. "The New Institutionalism@ Organizational Factors in Political Life." American Political Science Review 78: 734-49.
McCormick, James M. 1992. American Foreign Policy and Process. Itasca, IL: Peacock. O'Connor, Karen, and Lee Epstein. 1983. "The Rise of Conservative Interest Group Litigation." Journal of Politics 45: 479-89.
Petersen, Paul E. 1994. "The President's Dominance in Foreign Policy Making." Political Science Quarterly 109: 215-34.
Provine, Doris Marie. 1980. Case Selection in the United States Supreme Court. Chicago: University of Chicago Press.
Rossiter, Clinton. 1956. The American Presidency. New York: Harcourt, Brace. Schlesinger, Arthur M. 1973. The Imperial Presidency. Boston: Houghton Mifflin.
Segal, Jeffrey A. 1988. "Amicus Curiae Briefs by the Solicitor General during the Warren and Burger Courts." Western Political Quarterly 41: 135-44.
_. 1990. "Supreme Court Support for the Solicitor General: The Effect of Presidential Appointments." Western Political Quarterly 41: 135-44.
Segal, Jeffrey A., and Cheryl D. Reedy. 1988. "The Supreme Court and Sex Discrimination: The Role of the Solicitor General." Western Political Quarterly 41: 553-68.
Segal, Jeffrey A., and Harold Spaeth. 1993. The Supreme Court and the Attitudinal Model. Cambridge: Cambridge University Press.
Silverstein, Gordon. 1997. Imbalance of Powers: Constitutional Interpretation and the Making of American Foreign Policy. New York: Oxford University Press.
Smith, Rogers M. 1988. "Political Jurisprudence, The 'New Institutionalism' and the Future of Public Law." American Political Science Review 82: 89-108. Ulmer, S. Sidney 1978. "Selecting Cases for Supreme Court Review: An Under
dog Model." American Political Science Review 72: 902-10. Wildavsky, Aaron. 1966. "The Two Presidencies." Transaction 7-14.
Yates, Jeff, and Andrew Whitford. 1998. "Presidential Power and the United States Supreme Court." Political Research Quarterly 51: 539-50.
KIMI LYNN KING AND JAMES MEERNIK UNivERsiTy OF NORTH TExAs
Received: September 21, 1998 Accepted: August 23, 1999 email@example.com firstname.lastname@example.org
After selecting the set of Supreme Court cases with implications for U.S. foreign policy, a content analysis of the text was conducted to code the independent variables. Each Supreme Court opinion was Shepardized on Westlaw; and every single, available, lower court opinion was systematically examined as well to generate two separate sets of issue and actor codes. The following coding rules were established and relied upon for determining the variable measures.
Issue Areas (War Power Treaty Power, Recognition Power, and Civil Liberties)
Because the Supreme Court rarely certifies more than two to three questions on review or in original jurisdiction cases and because lower courts may address a range of issues, we obviously did not rely on the court having specifically indicated that the question presented involved a foreign or domestic issue area. Instead we relied upon the Westlaw Key Number Service, available headnotes, syllabi, and case summaries. We used the exact same method for determining the issues in the lower court opinions as well.
The Westlaw Key Number System provides one way to reassure the reader that we were stringent in our criteria for selecting cases and for standardizing the research/coding process. Each of our 347 cases had headnotes that indicated the topic of law discussed by the court. Of the Over 400 Key numbers in the Westlaw series, our cases included a headnote in at least one of the following key numbers areas:
Administrative Law and Procedure, Admiralty, Aliens, Ambassadors and Consuls, Appeal and Error, Attorney General, Citizens, Civil Rights, Commerce, Common Law, Constitutional Law, Corporations, Counties, Criminal Law, Customs and Duties, District of Columbia, Extradition and Detainers, Federal Civil Procedure, Federal Courts, Insurrection and Sedition, International Law, Judgment, Labor Relations, Libel and Slander, Mandamus, Military Justice, Navigable Waters, Neutrality Laws, Oath, Officers and Public Employees, Public Lands, Public Utilities, Seaman, Searches and Seizure, States, Statutes, Territories, Trade Regulation, Treason, United States, War and National Emergency, and Water and Water Courses.
To determine the list of issue codes, we relied upon doctrinal analyses of U.S. foreign policy (e.g., Adler and George 1996; Koh 1990). We coded for up to four issues in every opinion. Each case was coded according to whether the headnotes, syllabi, or case summaries indicated that the appropriate issue was discussed. First, cases involving War/National emergency powers/Commanderin-Chief were coded 1. Other cases were coded 0. Second, cases addressing treaty power and authority were coded 1. Other cases were coded 0. Third, cases regarding recognition powers (receiving foreign ambassadors/recognizing appropriate sovereign nations) were coded 1. Other cases were coded 0. The same coding scheme was applied to cases involving one of the general civil liberties (including speech, press, religion, due process, travel, habeas corpus, search and seizure, and the Ninth Amendment). Other cases were coded 0.
To determine whether federal supremacy was involved in the litigation, we relied upon the listing of parties, legal counsel, and the text of the opinion itself. Additionally, we wanted to use the more stringent coding criteria that the state's interests must be opposed to the national interest (rather than the federal govemment being aligned with one or more of the states). A case was coded I if a state was a named a party to the lawsuit, if the state had been granted intervenor status, or if a state's Attorney General was listed among the opposing counsel and the state(s) were opposed by the executive branch. Other cases were coded 0. Foreign Actor
In contrast to federal supremacy, the determination regarding foreign actor status was not as straightforward. We relied upon the listing of the parties and the text of the opinion itself. Frequently the court would discuss the litigant's status in either the summary or the opening paragraph setting out the facts of the case. This was not always the case-and in a rare number of instances we relied upon the full text of the case. Opinions were coded I if the opposing litigants represented "foreign," "international," or "non-domestic" entities whether public or private. Other opinions were coded 0.
To determine whether the court addressed the executives authority we relied upon available headnotes, syllabi, and case summaries, as well as the text of the opinion itself. The LOCATE command in Westlaw was used with the search string execut!, commander, or resident!. We examined systematically all references within the opinion that might have addressed presidential power or authority. Consistent with Silverstein (1997), we required that the opinion (either Supreme Court or lower court) remain silent on the Executive's "power," "discretion," "ability," "capability," "duty," "responsibility," "approval," "recognition," "primacy," "right," "obligation," "competency," and "control." All cases which did address these concerns as they related to a Constitutional provision, statutory law, or administrative regulation were coded 1. Others were coded 0. Separation of Powers
To determine whether the case presented an issue involving the separation of powers between the Executive and legislative branches, we once again relied on available headnotes, syllabi, and case summaries, as well as the text of the opinion itself. The LOCATE command in Westlaw was used with the search string "execut!," "commander," "president!," "congress!," or "legislat!." We examined sytematically all references within the opinion that might have addressed presidential power or authority as it related to legislative power or authority. Consistent with Silverstein (1997), we required the more stringent criteria that the opinion (either Supreme Court or lower court) specifically discuss the President's power in relation to the legislative branch. That is, both the President and the Congress must have been discussed within the same paragraph of the opin
ion, and within that same paragraph the author must have discussed the "power," "discretion," "ability," "capability," "duty," "responsibility," "approval," "recognition," "primacy," "right," "obligation," "competency," and "control" of either the executive or the legislative branch. In essence, the court is raising the issue explicitly noting one of two things. Either the opinion must note that the Executive is interfering with a power reserved to Congress, or else, that the Congress is interfering with a power reserved to the executive branch. All cases where the opinion indicates that either the Presidency or the Congress is either taking or delegating powers that are not allowed to it or reside within the other branch are coded 1. Others are coded 0.
Presidential Power Cases
We identified cases pertaining to presidential power (those analyzed in Table 3) by means of the following criteria. The reference in the Supreme Court opinion must have related to the Executive/President and to one of the following terms: "power," "inherent power," "authority," "inherent authority" "discretion,"
"inherent discretion," "ability," "inherent ability," "capability," "capacity," "duty," "responsibility," "approval," "recognition," and "primacy" and "right." The reference may have also related the Executive/President specifically to one of the following phrases: "enter into obligation(s)", "is obligated," "has the discretionary authority," "maintain(s) authority," "had the competency," "ability to control," "would enable," "establish the construction of laws," "may prescribe rules," and "formulated policy." We did not use the referent if it related to either the executive branch, or the presidential administration. The foreign policy reference must have specifically referred to the Executive/President as an individual. In per curiam cases, we also relied on concurring opinions.…