Authors: Nancy Scherer and Banks Miller
Publication: Political Research Quarterly 62: 366-378
Original source: fed_soc_final.pdf
Conversion note: this Markdown was generated by text extraction from the hosted file. Review equations, tables, notes, and citations before treating it as a final web copy.
Political Research Quarterly V olume XX Number X Month XXXX xx-xx © 2008 University of Utah 10.1177/1065912908317030 http://prq.sagepub.com hosted at http://online.sagepub.com 1 The Federalist Society’s Influence on the Federal Judiciary Nancy Scherer Wellesley College Banks Miller The Ohio State University Only twenty-five years after its founding, the Federalist Society today boasts a nationwide membership including renowned attorneys, politicians, policy makers, and jurists. Although the Society maintains that it is not a political organization, liberal political activists claim the Society has long pursued an ambitious—and extremely conservative—political agenda. In this article we ask: do members of the Federalist Society decide cases in a more conservative manner than other nonmember jurists? Using data on decision making in the U.S. Courts of Appeals, we find Federalist Society members are significantly more conservative than nonmembers and examine the long-term implications of our study. Keywords: judges; courts; decision making; appointments T he first chapter of the Federalist Society was founded at Yale Law School in 1980, and the following year, a second chapter began at the University of Chicago Law School (Bossert 1997; Hicks 2006). By 1982, the Federalist Society was legally established as a nonprofit corporation and thereafter became a national organization (McIntosh 2003). From its inception, the Federalist Society sought to provide a counterbalance to the “liberal jurisprudence” that prevailed in the nation’s law schools at that time, most notably by developing a method of constitutional interpretation now known as textualism (alternatively referred to as originalism). As the Society’s Executive Director, Eugene Meyer, explains textualism: If a word meant “x” when the Constitution was passed, and it means “y” today . . . you presumably want to stick with the meaning it had before. But for the most part, [textualism requires that] you interpret what the meaning [of the Constitution] was [when passed] based on what the meaning was in the past, and you pay attention to changes [in that meaning] over time. (Meyer 2002) The Society has also long focused attention on the proper role of federal court judges in our democracy. They contend that judges should practice “judicial restraint”; unelected jurists should not interject their personal policy preferences when interpreting the Constitution: Is the court interpreting the text and meaning of the Constitution? If it is, and they [the judges or justices] are doing the best they can do . . . their judgment might be off, but there is not a structural problem. If the court is saying, “gee, we don’t like the direction policy is going in this country [and w]e want to change the direction of policy . . .” that is not a proper role for the courts. (Meyer 2002) It has also been part of the Federalist Society’s central mission “to present their ideas [about originalism] more articulately . . . [which] could cause others to listen to these views more attentively, and, perhaps ultimately, to question some of the liberal positions which are being presented as the law” (Proposal to Form a National Conservative Legal Authors’ Note: The authors wish to thank Lawrence Baum, Wendy Martinek, and Rachel McGuire for their extremely helpful comments and advice on earlier drafts of this manuscript. We also wish to thank Amy Steigerwalt for her role in conducting some of the interviews cited herein. Finally, we wish to thank Steven Teles for allowing us to read a draft of his recently published (and extremely informative) book on the conservative legal movement. Political Research Quarterly OnlineFirst, published on May 1, 2008 as doi:10.1177/1065912908317030 Copyright 2008 by University of Utah.
Organization 1982, quoted in Teles 2008, 140 and 314). In short, the Society made plain as early as 1982 that it hopes to “transition” nonbelievers—lawyers who support the method of constitutional interpretation that sees the document as “living” and “evolving”—into committed supporters of originalism (Teles 2008). Because the Society’s positions on both constitutional interpretation and the role of unelected judges were deemed as viable legal arguments to further conservative political policy goals, the Society soon found itself important allies in Republican presidential administrations. Society Members Gain Influential Positions in the Executive and Judicial Branches The association between the Federalist Society and the Executive Branch began during the Reagan administration when Michael Horowitz, Reagan’s Director of the Office of Management and Budget, contacted the Society’s founding members and began introducing them to key people in the Reagan administration (Teles 2008). Edwin Meese—Counselor to the President (1981–1985) and Attorney General (1985–1988)—was another early supporter of the organization (Shapiro 1998; Teles 2008). Meese also urged all lawyers at Justice to become affiliated with the Federalist Society (Landay 2000). To this end, in 1985, the Federalist Society created its lawyers’ division for practicing attorneys, law professors, and judges. Meese then began elevating Federalist Society members to positions of importance in the Reagan Justice Department. By 1986, all twelve of the Assistant Attorneys General in the Justice Department were tied to the Federalist Society (Lewis 1991). 1 Notably, the three original founders of the Society, Steven Calabresi (from Yale) and David McIntosh and Lee Liberman (from Chicago), held key policy making positions in the Reagan Justice Department (New York Times 1986). With a Justice Department replete with Federalist Society supporters, it was not long before the Reagan administration began selecting Society allies for important seats on the federal bench, including Antonin Scalia to the Supreme Court. Reagan made all three of the Society’s original faculty advisors federal court judges. 2 The Society continued to wield significant influence in the George H. W. Bush (“Bush [41]”) Justice Department. Federalist Society members C. Boyden Gray (Counsel to the President) and Lee Liberman (Deputy Counsel to the President) were reportedly very involved in choosing Bush (41)’s circuit court judges and Supreme Court justices (Miner 1992). This period thus saw a second Society member appointed to the Supreme Court (Clarence Thomas) and eight more members nominated to the courts of appeals (though three were not confirmed). Set forth in Table 1 is a list of all Federalist Society members nominated by presidents Bush (41) and George W. Bush (“Bush [43]”) to the Supreme Court and the U.S. Courts of Appeals. 3 After Clinton was elected in 1992, the Federalist Society became a “Justice Department in exile” (Neas, quoted in Bossert 1997). However, during this 2 Political Research Quarterly
Federalist Society Members Nominated to U.S. Supreme Court and the U.S. Courts of Appeals by Administration, 1989–2005 Membership Year First Listed on Name Court Nominated Questionnaire Bush (41) Samuel Alito 3rd Circuit 1990 Yes Lillian Bevier+ 4th Circuit 1991 Not available Dennis Jacobs 2nd Circuit 1992 Yes Francis Keating + 10th Circuit 1991 Not available Michael Luttig 4th Circuit 1991 No Sheldon Plager Federal Circuit 1989 Yes John Roberts + D.C. Circuit 1992 Not available Clarence Thomas D.C. Circuit 1989 No Clarence Thomas Supreme Court 1991 No Bush (43) Samuel Alito Supreme Court 2005 Yes Carlos Bea 9th Circuit 2003 Yes Janice Brown ++ D.C. Circuit 2003 Yes Jay Bybee++ 9th Circuit 2001 Yes Michael Chertoff 3rd Circuit 2003 No Edith Clement 5th Circuit 2001 Yes Steven Collonton 8th Circuit 2003 Yes Deborah Cook ++ 6th Circuit 2001 No Miguel Estrada+ D.C. Circuit 2001 Yes Thomas Griffith++ D.C. Circuit 2004 No Raymond Gruender 8th Circuit 2003 Yes Harris Hartz 10th Circuit 2001 Yes Brett Kavanaugh ++ D.C. Circuit 2003 Yes Carolyn Kuhl+ 9th Circuit 2001 No Michael McConnell 10th Circuit 2001 Yes David McKeague ++ 6th Circuit 2001 Yes Priscilla Owen++ 5th Circuit 2001 No William Pryor++ 11th Circuit 2003 Yes John Roberts++ D.C. Circuit 2001 No John Roberts Supreme Court 2005 No John Rogers 6th Circuit 2001 Yes Henry Saad + 6th Circuit 2001 Yes Jeffrey Sutton++ 6th Circuit 2001 Yes Diane Sykes 7th Circuit 2003 Yes Timothy Tymkovich ++ 10th Circuit 2001 Yes +Not confirmed. ++Confirmed in subsequent congressional session.
period, Society members continued to network, and young lawyers affiliated with the Society were awarded prominent clerkships with conservative federal judges also affiliated with the Society (Bossert 1997). With the election of Bush (43) in 2000, however, Society members would once again gain prominent positions in the federal government, including many high-ranking current and former Justice Department officials. 4 But, what sets the Bush (43) administration apart from prior Republican administrations is its reliance on the Federalist Society to fill a majority of its appointments to the Supreme Court and courts of appeals. As we try to monitor the legal DNA of President [W.] Bush’s [judicial] nominees, we find repeatedly the Federalist Society chromosome. Why is it that membership in the Federalist Society has become the secret handshake of the [W.] Bush nominees for the federal court? (Benac 2005, 2, quoting Durbin) 5 Both of Bush (43)’s Supreme Court nominees, and more than two-thirds of Bush (43)’s nominations to the courts of appeals made in his first term are Federalist Society members (see Table 2). Liberal Interest Groups Accuse the Federalist Society of a Right Wing Cabal Notwithstanding the significant influence the Society and its members have wielded in three straight Republican administrations in terms of staffing the Justice Department, the West Wing, and the federal bench, the organization has always maintained that it is a nonpolitical organization. Thus, for example, the Society has made a point of refraining qua organization from taking public stands on particular legal cases or endorsing the appointment of specific people to the bench (Bossert 1997). But, liberal activists counter, the Society nonetheless seeks to influence legal decisions through a different method than that traditionally followed by interest groups: by encouraging Republican lawyers to join the organization; indoctrinating them to adopt the Society’s conservative method of constitutional interpretation known as textualism; and then seeing that Federalist Society members are strategically placed in positions of influence in the Justice Department and on the federal bench. In support of their argument, liberal activists are quick to point to a document prepared by Society affiliates serving in the Reagan Justice Department (Cavendish 2002; Gandy 2002; Neas 2002). 6 Specifically, in 1988, the Reagan Justice Department issued the Guidelines on Constitutional Litigation (the “Guidelines”). The Guidelines were intended to act as a roadmap for U.S. Attorneys litigating cases in federal district courts (Department of Justice 1988, 1). The Guidelines were not mere suggestions, but instead, “should presumptively be followed” (Department of Justice 1988, 1). As a primary tenet of constitutional interpretation, the Guidelines instructed U.S. Attorneys to advance, whenever possible, constitutional arguments using textualist interpretations of the Constitution. Prosecutors’ primary duty at this stage of the litigation was to “educate” lower court judges on the theory of textualism and to lay the groundwork for strong appellate arguments (Department of Justice 1988, 3). The Guidelines also set forth specific areas of law for which these textualist arguments were most relevant, including cases involving civil liberties for criminal defendants, federalism, civil rights for minorities, and privacy rights. To this end, the Guidelines specifically instructed prosecutors to argue in legal briefs for the reversal of some of the Supreme Court’s most seminal decisions of the twentieth century, including Roe v. Wade, Miranda v. Arizona, and Wickard v. Filburn (Department of Justice 1988, 54, 82, 86). 7 In conformance with the Guidelines’ vision—and consistent with the Society’s central tenet to advance a textualist interpretation of the Constitution— Republican administrations have repeatedly tried to overturn longstanding Supreme Court precedents that conservatives often characterize as “liberal” decisions made by “activist” justices and judges. For example, in Planned Parenthood v. Casey(1992), the Bush (41) administration sought to overturn Roe and in Dickerson v. U.S. (2000), it sought to overturn Miranda. In both of these cases, the Federalist Society members then sitting on the Supreme Court (Scalia in Casey; Scalia and Thomas in Dickerson) voted to overturn Roe and Miranda. 8 Liberal interest group leaders also contend that, with the increasing number of Federalist Society members on the federal courts, the organization’s goals—first articulated during the Reagan administration—are well within reach: Their [the Federalist Society members’] goal was to change the way that the law is made and enforced in this country. To make a sea change Scherer, Miller / The Federalist Society’s Influence 3
in the way everything was done. And they have done it little by little. They started small. And they have in the last 20 years, I think, already accomplished an enormous amount of what they wanted to do and it has gone completely under the radar. (Gandy 2002) Other liberal activists concur, and fear that, ultimately, the Society will gain enough clout to put constitutional interpretation “back where it was before 1937” (Neas 2002). Others worry specifically about the future of civil rights laws passed in the 1960s (Gandy 2002; see also Aron 2002; Cavendish 2002). 9 In short, liberal activists believe that the Society is seeking to “accomplish in the courts what Republicans can’t achieve politically” (Gandy 2002; Neas 2002)—that is, to overturn progressive Supreme Court precedent and congressional legislation through the appointment of extremely conservative federal court judges. Hypothesis, Data, and Method We now subject to empirical analysis the claim that Federalist Society members sitting on the federal bench are more conservative than other jurists, including other Republican jurists: Judges who are affiliated with the Federalist Society are more conservative in their judicial decision making behavior than judges who are not affiliated with the Federalist Society. We test this hypothesis using data on decision making behavior of judges who sit on the U.S. Courts of Appeals. As the level of the federal judicial hierarchy with the most Federalist Society members, data on decision making in the courts of appeals provides the best opportunity to test our hypothesis. We focus on two separate legal issues to test our hypothesis, one criminal and one civil: (1) does a search alleged to be illegal under the Fourth Amendment require that incriminating evidence be suppressed at trial; and (2) should a congressional statute be declared unconstitutional because Congress has overstepped its authority to the detriment of a state’s sovereignty in violation of the Tenth or Eleventh Amendments? These issues were chosen for two critical reasons. First, while one of these issues, states’ rights, is inextricably intertwined with the Society’s core mission, the other, law and order, has been an issue embraced by the Republican Party (including Republican judges) since the Nixon administration (Scherer 2005). Second, both legal issues present a judge with clear choices between liberal and conservative outcomes. In search and seizure cases, the judge must either suppress evidence that incriminates the defendant so as to safeguard core constitutional rights (the liberal position) or deny defendant’s claim of a constitutional violation to maintain “law and order” (the conservative position). In the states’ rights area, the judge must choose between curtailing the power of the federal government (the conservative position) and expanding it (the liberal position). The unit of analysis is each judge’s vote (not the ultimate holding of the case), rendered on a threejudge or en banc appellate panel in all cases meeting certain criteria set out below. 10 The cases included in this study are the entire universe of “nonconsensual” decisions (defined below) in the two selected issue areas rendered by the United States Courts of Appeals, including all eleven numbered circuits, the District of Columbia Circuit, and the Federal Circuit. For the search and seizure data we use all cases decided between January 1, 1994, and December 31, 2005. For the states’ rights data, we use all cases decided from January 1, 1996, through December 31, 2006. Included in the analysis are the votes of circuit court judges in active service and circuit court judges with senior status. District court judges sitting by designation were excluded. Not only are these judges not necessarily fungible with appellate judges (Brudney and Distlear 2001), but also, there is insufficient newspaper reporting on these appointees to code the Federalist Society membership variable as described below. We include the votes of judges appointed by Presidents Bush (41), Clinton, and Bush (43). 11 The analysis focuses specifically on “nonconsensual” decisions rendered by the courts of appeals in the relevant time frame. To be considered a “nonconsensual” decision, one of two conditions must be met: (1) the appellate panel—be it a regular panel or an en banc panel—rendered a split decision (i.e., there was at least one dissenting vote against the majority ruling); or (2) the appellate panel, though unanimous in its own decisions, reversed or vacated the decision of the district court judge below. 12 We identified the cases comprising these two data sets from a series of comprehensive searches on the electronic database WESTLAW. Included in the data are decisions officially designated by the court for publication in the Federal Reporter,as well as decisions not designated by the court for official publication but reprinted in full on the WESTLAW database. 13 4 Political Research Quarterly
We constructed two models (one for each data set), employing a number of control variables in addition to the major explanatory variable, Federalist Society membership. Dependent Variables The dependent variable in the states’ rights data set is the vote of each individual judge either agreeing or disagreeing with the plaintiff that the state government’s rights prevail over the federal government’s rights. In the Tenth Amendment cases, this would mean a vote agreeing or disagreeing with the states’ rights advocate that the federal government has exceeded its power under Article I, Section 8 of the Constitution. In the Eleventh Amendment cases, this would mean a vote agreeing or disagreeing with the state that it is immune from suit, and that the federal government has not properly abrogated state sovereign immunity in the challenged congressional statute. Collectively, these two categories of cases are referred to throughout as “states’ rights” cases. The dependent variable in the search and seizure data set is the vote of each individual judge either agreeing or disagreeing with the criminal defendant that, by virtue of the government’s violation of his or her Fourth Amendment rights, incriminating evidence of the defendant’s guilt must be suppressed at trial. Independent Variables Federalist Society membership . Because the Federalist Society is a loose confederation of law students, lawyers, and judges, it is difficult to pin down precisely who on the federal bench is officially a dues-paying “member” of the Society as opposed to someone merely affiliated with the organization and its policy stances (Peterson 2005). Membership lists are carefully guarded. Accordingly, we code this variable two ways, each using a different definition of Federalist Society Affiliate. The first definition is a judge who listed that he or she was a member of the organization on his or her questionnaire submitted to the Senate Judiciary Committee as part of the confirmation process. Questionnaires are available to the public for all nominees from 1989 to the present. The problem with this very narrow definition of Federalist Society member is that, increasingly, actual members of the Society have chosen to omit their affiliation with the organization on their questionnaires. This trend came to light during the confirmation hearings of Chief Justice John Roberts in 2005. Though his questionnaire stated nothing about being a member of the Federalist Society, the Washington Post uncovered that he was not only a member, but was a member of the Society’s leadership council (Lane 2005). Moreover, during the earlier days of the Society’s existence, when there were fewer practicing attorneys chapters, lawyers and law professors who were appointed to the bench may have had very close ties to the Federalist Society but technically were not “members.” Accordingly, such close affiliates may have omitted their ties to the organization on their questionnaire because they were not dues-paying “members.” To account for the under-reporting of Federalist Society membership on Judiciary Committee questionnaires, we alternatively code this variable by including not only those listing membership on their questionnaires, but also, judges who are identified in at least two separate newspaper articles as “members” of the Federalist Society. Nominees characterized as merely affiliated in some way with the Society were not counted as being members for purposes of this study. 14 Newspaper articles were identified through the Lexis/Nexis and Factiva databases. Because news of a court of appeals nomination is often of more local interest, rather than of national interest, it was necessary for us to search beyond what are considered national newspapers like the New York Times,the Wall Street Journal,and the Washington Post(though these newspapers are included in the two data bases we used for our searches). The states’ rights model also includes a number of variables that previous scholarship has identified as relevant to analyzing judicial voting in federalism cases (e.g., Collins 2006; Solberg and Lindquist 2006) or more generally in models involving judicial review (Sala and Spriggs 2004). These include the principal legal issue (Tenth versus Eleventh Amendment challenge); the ideological proximity of the challenged statute to the judge; the ideological proximity of challenged statute to Congress; the additional number of liberal amicus curiae briefs (versus conservative briefs); and the age of the statute. For the search and seizure model, we include a number of variables related to facts in search and seizure cases that previous scholarship has identified as impacting judicial decision making behavior (e.g., Segal 1983, 1986). These include place of the search, presence of a warrant, extent of the search (limited search such as a pat-down versus a fullblown search) and whether the search took place at an international border. 15 Scherer, Miller / The Federalist Society’s Influence 5
In both models, we also control for variables related to the judge’s background including political ideology, race, gender, circuit and whether the judge has taken senior status. We detail the coding of all independent variables in the online appendix. Results Both models are estimated using logit, as our dependent variables are measured dichotomously. Because some judges in the data sets vote multiple times, our models use robust standard errors to correct for potential heteroskedasticity and also cluster around each judge to account for possible nonindependence between observations. 16 Although dummy variables for each circuit were included in the models, we do not report these coefficients in the tables. The Seventh Circuit was used as the baseline circuit in the model because it is generally deemed neither a particularly liberal nor conservative circuit.
model. As expected, the Federalist Society coefficient is positive and statistically significant (using α = .05, one-tailed test). This is true even after controlling for political ideology (which is also statistically significant). As hypothesized, members of the Federalist Society do exhibit more conservative voting behavior than nonmembers. Turning to the rest of the states’ rights model, several other coefficients are statistically significant. A judge is less likely to vote for the states’ rights position if he is black; an Eleventh Amendment challenge is made; more pro-federal government amici briefs are filed; and the greater the distance the policy is from the current Congress. To facilitate interpretation of the logit coefficients we have calculated predicted probabilities of votes in favor of the states’ right litigant. We assume the following conditions: a white male judge in active service; deciding a Tenth Amendment case; with the same number of amici briefs filed on the pro- and anti–federal rights sides; and involving a federal statute that has moderate ideological distance from Congress. We present the pro–states’ rights vote by appointing president, ideology, and Federalist Society membership. These post-estimation probabilities are depicted in Figure 1. From Figure 1 we can see that the most liberal Clinton appointee has a 0.25 probability of voting for the state’s rights position, the median Clinton appointee a 0.37 probability, and the most conservative Clinton appointee a 0.48 probability. For Bush (41) appointees who are not Federalist Society members, the most liberal of this cohort has a 0.52 probability of voting for the conservative position, the median a 0.70 probability, and the most conservative a 0.72 probability. For Bush (43) judges who are not members of the Federalist Society, the most liberal of this cohort has a 0.59 probability of voting for the states’ rights position, the median judge a 0.66 probability, and the most conservative a 0.73 probability. Even Republican appointees not affiliated with the Society stand solidly to the right of Democratic appointees on the ideological spectrum when deciding states’ rights issues. For Federalist Society members, there is an even higher probability that the judge will support a textualist interpretation of the Tenth and Eleventh Amendments. In the Bush (41) administration, for a judge who is deemed less conservative by traditional judicial ideology measures, membership in the Society raises the probability of casting a conservative states’ rights vote from 0.52 to 0.88 (a difference of 0.36); for a judge whose ideology score places him at the median of this cohort, membership increases the probability of casting a conservative vote from 0.70 to 0.94 (a difference of 0.24); and for the most conservative among the cohort, membership increases 6 Political Research Quarterly
States’ Rights Model Nonconsensual Decisions, U.S. Courts of Appeals, 1996–2006++ Robust Standard β Error P Judicial Background Variables Federalist Society member 1.90 0.61 0.001 Ideology of judge 1.79 0.74 0.008 Black judge –1.05 0.58 0.036 Female judge –0.01 0.43 0.492 Senior judge 0.89 0.84 0.143 Case Variables Eleventh Amendment –0.63 0.38 0.047 Additional liberal amici –0.50 0.14 0.000 Judge-policy proximity 0.48 0.50 0.168 Congress-policy proximity –7.22 3.22 0.012 Age of statute 0.00 0.01 0.407 Constant –1.48 1.06 0.082 Note: (pro-states’ rights vote coded 1; anti-states’ rights coded 0). N = 229 Chi-squared = 68.51; significant at .000 Pseudo R-squared = .214 P values are one-tailed. +Coefficients for circuit control variables are omitted. Seventh Circuit was used as the baseline in the model.
the likelihood of a pro–states’ rights vote from 0.72 to 0.95 (a difference of 0.23). Results for Society members appointed by Bush (43) are very similar to those for Bush (41) appointees. For the Bush (43) appointee with the most liberal ideology score, Society membership increases the likelihood of a conservative federalism vote from 0.59 to 0.90 (a difference of 0.31); the Bush (43) appointee with the median ideology score from 0.66 to 0.93 (a difference of 0.27); and the most conservative from 0.73 to 0.95 (difference of 0.22). With respect to the search and seizure cases we ran two separate models, one using our broad definition of Federalist Society member (identified through questionnaires and newspaper articles) and one the narrower definition (questionnaires only). Because there was no substantive difference between the results of the two models, we present only the results of the model using the broader definition of Federalist Society membership (see Table 3). 17 Like the states’ rights model, we again find the coefficient for Federalist Society membership is statistically significant, even after controlling for ideology (which is slightly higher than the .05 sigificance level used in this study). As hypothesized, Society members are more likely to uphold the government’s search as consistent with the Fourth Amendment (the conservative position) than a nonmember. We also find that black judges are less likely than white judges to rule in favor of the government when a search is challenged in court, and senior judges more likely than active judges to rule in the government’s favor. Moreover, although slightly above the p = 0.05 level of statistical significance (one-tailed test), searches made with a warrant are more likely to be upheld than warrantless searches. As we did with the states’ rights model, we also calculated the predicted probability of a conservative search and seizure vote to facilitate interpretation of the logit coefficients. We assume a full-blown search of a home without a warrant and decided by a white male Scherer, Miller / The Federalist Society’s Influence 7
Predicted Probabilities of Vote in Favor of States’ Rights Litigant by Appointing President, Ideology, and Federalist Society Membership++ +Assumes a white male judge in active service, deciding a Tenth Amendment case with an equal number of amici briefs on both side s, and involving a statute that is moderately distant from the median member of Congress.
judge still in active service. These post-estimation probabilities are depicted in Figure 2. As illustrated in Figure 2, the most liberal Clinton appointee has a 0.31 probability of voting against a criminal defendant in a search and seizure case, the ideologically moderate Clinton appointee a 0.34 probability, and the most conservative a 0.37 probability. For the most liberal Bush (41) appointee who is not a member of the Federalist Society, there is 0.38 probability of a vote against a defendant, a 0.42 probability for a judge who lies in the middle of the ideological spectrum for this presidential cohort, and a 0.44 probability for the most conservative among this cohort. For non-member Bush (43) appointees, the most liberal of the cohort has a 0.40 probability of casting a conservative criminal decision, the median judge a 0.41 probability, and the most conservative a 0.43 probability. More importantly, across the board, Federalist Society membership substantially increases the likelihood that a Republican judge will vote against a criminal defendant on a suppression motion turning on the Fourth Amendment. In the Bush (41) administration, for a judge who is deemed less conservative by traditional judicial ideology measures, membership in the Society raises the probability of casting a conservative criminal vote from 0.38 to 0.75 (a difference of 0.37); for a judge whose ideology score places him at the median of this cohort, membership increases the probability of casting a conservative vote from 0.42 to 0.78 (a difference of 0.36); for the most conservative judge in the cohort, membership increases the likelihood of a conservative vote from 0.44 to 0.80 (a difference of 0.36). For Bush (43) appointees, Society membership increases the likelihood of a conservative vote from 0.40 to 0.77 (a difference of 0.37) for the most liberal of the cohort; from 0.41 to 0.78 (a difference of 0.37) for the median judge in the cohort; and from 0.43 to .79 (a difference of 0.36) for the judge with the most conservative ideological score in the cohort. Why Is the Society Influencing Decision Making? Without question, our results demonstrate that Federalist Society membership has a statistically significant and substantively large impact on judicial decision-making behavior on the U.S. Courts of Appeals. This is true not only with respect to legal issues inextricably intertwined with the Federalist Society, such as states’ rights, but also with issues long associated with the Republican Party writ large. We now consider why the Federalist Society may be influencing judicial decision making. There are two possible explanations as to why Federalist Society membership increases the likelihood of a conservative vote. One hypothesis is that the Society attracts its members from among those Republican lawyers and law students who already lie on the conservative end of the ideological continuum; this theory we term the “attraction” theory. Pursuant to the attraction theory, a variable controlling for Society membership may be conceived of as simply measuring a judge’s political ideology. A second hypothesis is that the Society converts Republican lawyers who join its ranks to adopting the originalist method of constitutional interpretation. In other words, membership in the Society somehow socializes lawyers who may not ascribe to the originalist mode of constitutional interpretation when they join the organization—instead ascribing to the “living and evolving” constitutional paradigm—in such a manner that, over time, they are transformed into proponents of originalism; this theory is termed the “conversion” theory. 8 Political Research Quarterly
Search and Seizure Model Nonconsensual Decisions, U.S. Courts of Appeals, 1994–2005++ Robust Standard β Error P Judicial Background Variables Federalist Society member 1.60 0.38 0.000 Ideology of Judge 0.47 0.30 0.055 Black judge –1.16 0.31 0.000 Female judge 0.04 0.21 0.420 Senior judge 0.92 0.54 0.045 Case Variables Place of Search Home Baseline — — Car 0.42 0.22 0.026 Luggage 0.09 0.29 0.372 Person 0.15 0.24 0.261 Warrant obtained 0.44 0.28 0.056 Limited search 0.01 0.21 0.484 Border search –0.29 0.41 0.240 Additional liberal amici 0.23 0.35 0.252 Constant –0.80 0.43 0.031 Note: (vote against defendant coded 1; vote against defendant coded 0). N = 923 Chi-squared = 128.31; significant at .000 Pseudo R-squared = .108 P values are one-tailed. +Coefficients for circuit control variables are omitted. Seventh Circuit was used as the baseline in the model.
Unfortunately, given our data, it is impossible to reject the null hypothesis of either theory. At first glance, it may seem that we could rule out the attraction theory because our models control separately for Society membership and the judge’s political ideology, and membership is still statistically significant. One could argue that if Society membership were measuring political ideology, then we would expect that this coefficient would not be statistically significant when political ideology is controlled for separately. However, it is just as plausible that the political ideology measure we use—and that is now the standard measure in studies of lower court decision-making behavior (Giles, Hettinger, and Pepper 2001)—simply does not accurately capture the political ideology of the new breed of Republican judge which includes members of the Federalist Society. This is because the Giles et al. measure is predicated exclusively on traditional notions of what it means to be a Republican judge, measures developed prior to the predominance of Society members in the Republican cohort of judges on the federal bench. 18 Thus, in our models, it may be that the membership and ideology measures are both capturing political ideology. Our data also presents limitations in making firm conclusions about the conversion theory. On the one hand, our finding that Society members are almost twice as likely as other Republican judges to render a conservative vote on a search and seizure motion seems to refute the notion that the Society is “converting” otherwise traditional Republicans to their way of thinking. This is because of the fact that such a “law and order” approach to criminal civil liberties cases has been a mainstay of Republican judicial selection since the Nixon administration thus predating formation of the Society (Goldman 1997, 198). In short, Republican lawyers would not need the Federalist Society to introduce them to the idea that there is a competing approach to Fourth Amendment interpretation from that set down by Warren Court Scherer, Miller / The Federalist Society’s Influence 9
Predicted Probabilities of Vote against Criminal Defendant by Appointing President, Ideology, and Federalist Society Membership for the Search and Seizure Model++ +Assumes a white male judge in active service, deciding a case involving a full-blown search of a home with no warrant.
precedents. Thus, the fact that Federalist Society members are significantly more likely to adopt an originalist reading of the Fourth Amendment than nonmember Republicans suggests that the Society is simply attracting the most conservative lawyers in the Republican Party—the same type of lawyers that Nixon was trying to identify in declaring a “law and order” litmus test for his judicial appointments. But, to reject the null hypothesis (that the Society is not converting members), we would optimally need temporal data on judicial decision making—i.e., decisions made before and after particular judges joined the Society. More conservative voting behavior by a judge after joining the organization would be compelling evidence that the Society converts members into more conservative jurists; the absence of change would suggest that conversion is not at work. Such a study, however, is not possible using legal decisions given that Society-affiliated judges are chosen by Republican presidents specifically because they are already members of the organization and, as such, can be counted on to decide cases consistent with the president’s conservative policy agenda. One possible study that may be able to overcome this hurdle—and, critically, a research design not reliant on legal decisions—would be to conduct interviews with law students who are members of the Federalist Society and explore, through qualitative methods, the reasons why they joined the organization and how membership may have impacted their views on constitutional interpretation. 19 Long-Term Implications of This Study Notwithstanding the fact that no definitive conclusion can be drawn on why the Federalist Society influences judicial decision making, there are still important implications suggested by our central empirical finding. First, this study brings into serious question the method that political scientists are currently using to measure the political ideology of lower court judges. That method imputes senators’ and presidents’ Poole–Rosenthal ideology scores (Poole and Rosenthal 1997)—based on their public voting records (for congressmen) and public stances on congressional statutes (for presidents)—to lower court judges who have never taken a stand on these bills. Moreover, to the extent that these Republican senators and presidents are not affiliated with the Federalist Society, we now have clear evidence that the elected officials’ ideology scores are underestimating the conservativeness of Republican-appointed judges who are affiliated with the Society. 20 Future studies of judicial voting behavior should include control variables for Federalist Society membership (as well as race and gender) in addition to the Poole-Rosenthal scores. Second, this study demonstrates that there is now at work a new model for interest groups to follow in pursuit of their goal of influencing law and legal policy. Heretofore, studies of interest group influence on legal decision making have centered on three distinct strategies: sponsoring or financing federal court litigation (e.g., Epstein 1985; Schlozman and Tierney 1986); filing amicus curiae briefs in other parties’ federal court cases (e.g., Caldeira and Wright 1988); and lobbying the president and senators concerning judicial appointments to the Supreme Court (Caldeira and Wright 1998) and the lower federal courts (Bell 2002; Scherer 2005). The Federalist Society, in contrast, has forged a new path. It has sought to influence decision making by educating young lawyers to reject the conventional method of constitutional interpretation taught in the nation’s law schools, and instead, adopt originalism as the only correct method of interpretation. In short, through their decades-long intellectual and networking efforts, the Society has made originalism a widely accepted alternative to “living constitutionalism” and, as such, has provided conservative lawyers, sitting judges, and future judges the legal tools they need to reach conservative outcomes in constitutional cases. In this way, the Society accomplishes what more traditional grass roots organizations do, not by lobbying elected officials or litigating in court, but rather, by educating Republican lawyers. Third, this study has serious implications on what types of judges we can expect future Republican presidents to appoint. Given how accurate and powerful a predictor Society membership is as to the likelihood that a potential nominee will have an extremely conservative voting record, it should be expected that future Republican presidents (at least those elected on a conservative platform) will increasingly turn to the ranks of the Federalist Society to nominate the majority of their federal court judges. We have already witnessed such a trend in the last two decades (see Table 1). Society membership, in essence, eliminates for a conservative Republican president the uncertainty surrounding a nominee’s ideology and thus allows that president to avoid the situation that Bush (41) faced when it became apparent that his first Supreme Court appointee, David Souter, was not as conservative as originally thought, or even conservative at all. Fourth, and related to the last point, as Republican presidents increasingly turn to the Federalist Society to 10 Political Research Quarterly
make their federal court nominations, we should in turn expect Democratic senators to oppose Republican federal court nominations with greater frequency. Democratic senators come under a lot of pressure from liberal interest groups to block confirmation of Society members who are viewed as hostile to progressive social issues. Accordingly, it is likely that the judicial confirmation process will become further polarized than it is today. Finally, with more Federalist Society members on the bench, we should also, increasingly, see written constitutional opinions focusing on the text and the original intent of the Framers, rather than on recent history and currently acceptable social norms as non- Society members have tended to do. Future research should focus on content analyses of the case law to track this change. Conclusion The results of this study confirm liberal interest groups’ claims that Republican-appointed judges tied to the Federalist Society are more conservative than judges who are not members. If the trend started by Bush (43)—making Society members the majority of appointments to the two highest levels of the federal judicial hierarchy—continues with future Republican presidents, the Federalist Society will arguably have gone a long way in realizing its original goal to change the way we, as Americans, interpret the Constitution. In short, there will be a sizable presence of Society members on the federal bench, and these jurists will interpret the Constitution according to the dictates of originalism, not the method which sees the Constitution as an evolving document. Notes 1. Given that the lawyers division of the Society was not started until 1985, and had few chapters in the mid-1980s, practicing attorneys could not formally become members of the Society during most of the Reagan administration. Instead, during this early period in the Society’s history, Justice Department officials and judges were characterized as “friends of” or “sympathetic to” the organization by virtue of speaking at one of the student chapters’ symposia (Carter 2001). In contrast, by 2001, when Bush (43) took office, there were an estimated twenty thousand practicing attorneys who were full-fledged “members” of the Society’s Lawyers Division (www.fed-soc.org/aboutus/ id.28/default.asp). 2. Antonin Scalia and Robert Bork were appointed to the D.C. Circuit appellate court and then the Supreme Court (though Bork ultimately was not confirmed to the high court), and Ralph Winter to the Second Circuit. 3. As detailed below, we used two alternative methods to measure Federalist Society membership. Both the broad definition (judges identified in newspaper articles as “members” of the Society) and the narrow definition (judges who identified themselves as members on the Judiciary Committee questionnaire) are noted on Table 1. 4. For example, members included Theodore Olson (former Solicitor General), Paul Clement (current Solicitor General), Larry Thompson (former Deputy Attorney General), and Viet Dinh (former Assistant Attorney General for Legal Policy). 5. Richard Durbin (D-IL) is a member of the Senate Judiciary Committee. 6. Elizabeth Cavendish is the former General Counsel of NARAL Pro-Choice America; Kim Gandy is the president of the National Organization for Women; Ralph Neas is the president of People for the American Way. 7. Roe held that the 14th Amendment’s Due Process Clause encompasses a substantive right to privacy, and that includes a woman’s right to terminate a pregnancy. Mirandaheld that the Fifth Amendment’s ban on self-incrimination—together with the Sixth Amendment’s requirement that all indigent criminal defendants be afforded appointed counsel—requires that the police inform people of these rights before interrogating them. Wickard held that a congressional statute prohibiting individuals from growing wheat for personal consumption was sufficiently related to interstate commerce so that it may be regulated by the federal government. 8. Current Supreme Court justice Samuel Alito, also a Society member, was on the Third Circuit panel that heard Casey; Alito voted to uphold Pennsylvania’s restrictive abortion statute in toto. 9. Nan Aron is the president of the Alliance for Justice. 10. Because of the possibility that en banc decision making differs from decision making on a three-judge panel, we ran alternative models for each data set analyzing only the votes made on three-judge panels. The coefficients for Society membership were substantively comparable to those in the models using all votes. 11. Although the cases analyzed contained votes of Presidents Eisenhower, Kennedy, Johnson, Nixon, Ford, and Carter, we excluded these votes because these appointments were made before the Federalist Society was founded. We exclude the votes of Reagan appointees because, as explained above (see note 1), membership opportunities in the Society for practicing attorneys during the Reagan administration were very limited. 12. In contrast to the Supreme Court, the circuit courts do not have discretionary dockets. As such, they decide over 99 percent of all federal appeals taken in the United States. Because the vast majority of these appeals are not “close” cases, but are taken so as to exhaust all possible legal avenues, there is generally only one clear decisional path available to the sitting judges. In other words, the case may be finally adjudicated— and full agreement reached by the appellate panel and the district court judge—without underlying political ideologies playing a role in the decision-making process. In short, fully consensual decisions in the lower federal courts are generally deemed to reflect decisions based on precedent, statutes, or facts (Goldman 1966; 1975). 13. There is some debate as to whether unpublished decisions bias a sample of cases (e.g., Songer 1988). Accordingly, we ran the analysis without these votes as well and the substantive results were largely the same. Scherer, Miller / The Federalist Society’s Influence 11
Aron, Nan. 2002. Interview with Nancy Scherer and Amy Steigerwalt. Tape recording. July 9. Washington, DC. Bell, Lauren Cohen. 2002. Warring factions: Interest groups, money and the new politics of senate confirmation. Columbus, OH: Ohio State University Press. Benac, Nancy. 2005. Group becoming must for some conservatives. Associated Press. July 18. Bossert, Rex. 1997. Conservative forum is a quiet power. National Law Journal. September 8, p. A1. Boyd, Christina L., Lee Epstein, and Andrew D. Martin. 2007. Untangling the causal effects of sex on judging. Paper presented at the Empirical Legal Studies Conference, New York. November. Brudney, James J., and Corey Ditslear. 2001. Designated diffidence: District court judges on the Courts of Appeals. Law and Society Review 35:565-606. Caldeira, Gregory A., and John R. Wright. 1998. Lobbying for justice: Organized interests, Supreme Court nominations, and United States Senate. American Journal of Political Science 42:499-523. ———. 1988. Organized interests and agenda setting in the U.S. Supreme Court. American Political Science Review82:1109-27. Carter, Terry. 2001. The in crowd: Conservatives who sought refuge in the Federalist Society gain clout. American Bar Association Journal 87:46-51. Cavendish, Elizabeth. 2002. Interview with Nancy Scherer and Amy Steigerwalt. Tape recording, July 11. Washington, DC. Collins, Paul. 2006. Towards an integrated model of Supreme Court’s federalism decision making. Paper presented at the Annual Meeting of the American Political Science Association, Philadelphia, Pennsylvania. August 31–September 1. Department of Justice, Office of Legal Policy. 1988. Guidelines on constitutional litigation . Government Printing Office, Washington, DC. Epstein, Lee. 1985. Conservatives in court.Knoxville, TN: University of Tennessee Press. Gandy, Kim. 2002. Interview with Nancy Scherer and Amy Steigerwalt. Tape recording. July 11. Washington, DC. Giles, Michael W., Virginia A. Hettinger, and Todd Peppers. 2001. Picking federal judges: A note on policy and partisan selection agendas. Political Research Quarterly 54:623-41. Goldman, Sheldon. 1966. V oting behavior on the United States Courts of Appeals, 1961–1964. American Political Science Review 60:374-83. ———. 1975. V oting behavior on the United States Courts of Appeals revisited. American Political Science Review 69: 491-506. ———. 1997. Picking federal judges: Lower court selection from Roosevelt through Reagan. New Haven, CT: Yale University Press. Hicks, George W. 2006. The conservative influence of the Federalist Society on the Harvard Law School student body. Harvard Journal of Law and Public Policy 2006:625-717. Johnson, Timothy, Paul Wahlbeck, and James Spriggs. 2006. The influence of oral arguments on the Supreme Court. American Political Science Review 100:99-113. Landay, Jerry. 2000. The Federalist Society: The conservative cabal that’s transforming American law. Washington Monthly. March. Available at www.washingtonmonthly.com/features/ 2000/0003.landay.html (accessed March 15, 2007). Landis, J. R., and G. G. Koch. 1977. The measurement of observative agreement for categorical data. Biometrics 33:159-74. Lane, Charles. 2005. Roberts listed in ’97-98 directory; court nominee said he has no memory of membership. Washington Post. July 25. Lewis, Neil. 1991. Conservative outsiders now at the hub of power. New York Times. March 29. ———. 2004. Mixed results for Bush in battle over judges. New York Times. October 22. McIntosh, David. 2003. Introduction to the Federalist Society 2002 symposium on law and truth. Harvard Journal for Law and Public Policy 2003:1-5. Meyer, Eugene. 2002. Telephone interview with Nancy Scherer. Tape recording. July 30. East Hampton, NY . Miner, Roger. 1992. Advice and consent in theory and practice. In Judges on judging, ed. David M. O’Brien. Washington, DC: C.Q. Press, 2004. Neas, Ralph. 2002. Interview with Nancy Scherer and Amy Steigerwalt. Tape recording. August 6. Washington, DC. 12 Political Research Quarterly
New York Times. 1986. The Federalist Society: Judge Scalia’s cheerleaders. July 23. Peterson, David. 2005. Legal society is strong in state. Minneapolis Star Tribune. August 1. Poole, Keith T., and Howard Rosenthal. 1997. Congress: A politicaleconomic history of roll-call voting.Oxford: Cambridge University Press. Sala, Brian R., and James F. Spriggs II. 2004. Designing tests of the Supreme Court and the separation of powers. Political Research Quarterly 57:197-208. Scherer, Nancy. 2004. Blacks on the bench. Political Science Quarterly 119:655-72. ———. 2005. Scoring points: Politicians, political activists and the lower federal court appointment process . Stanford, CA: Stanford University Press. Schlozman, Kay Lehman, and John T. Tierney. 1986. Organized interests and American democracy. New York: Harper and Row. Segal, Jeffery A. 1983. Predicting Supreme Court cases probabilistically: The search and seizure cases, 1962–1981. American Political Science Review 78:891-900. ———. 1986. Supreme Court justices as human decision makers: An individual-level analysis of search and seizure cases. Journal of Politics 48:938-55. Segal, Jeffery A., and Harold Spaeth. 2002. The Supreme Court and the attitudinal model revisited. New York: Cambridge University Press. Shapiro, Bruce. 1998. True believer. Salon.Available at www.salon .com/news/1998/07/03news.html (accessed March15, 2007). Smith, R. Jeffrey, and Dan Eggen. 2005. Gonzalez helped set the course for detainees. Washington Post. January 5. Solberg, Rorie Spill, and Stefanie Lindquist. 2006. Judicial review in the Rehnquist court: Activism, federalism and ideology. Journal of Empirical Legal Studies 3:237-61. Songer, Donald R. 1988. Nonpublication in the United States District Courts: Official criteria versus inferences from appellate review. Journal of Politics 50:206-15. Songer, Donald, Sue Davis, and Susan Haire. 1994. A reappraisal of diversification in the federal courts: Gender effects in the courts of appeals. Journal of Politics 56:425-39. Spaeth, Harold. 2006. United States Supreme Court Database. Available at http://www.as.uky.edu/polisci/ulmerproject/sctdata .htm (accessed Mar.15, 2007). Spriggs, James F., and Thomas Hansford. 2000. Measuring legal change: The reliability and validity of Shepard’s Citations. Political Research Quarterly 50:327-41. Teles, Steven. 2008. The rise of the conservative legal movement: Organizational mobilization and political competition . Princeton, NJ: Princeton University Press. Cases Cited Aguilar v. Texas,378 U.S. 108 (1964) Board of Trustees of the University of Alabama v. Garrett, 531 U.S. 356 (2001) Carroll v. United States, 267 U.S. 132 (1925) Chimel v. California, 395 U.S. 752 (1969) Coolidge v. New Hampshire, 403 U.S. 443 (1971) Davis v. Mississippi, 394 U.S. 721 (1969) Dickerson v. U.S., 530 U.S. 428 (2000) Florida Prepaid Postsecondary Education Expense Board v. College Savings Bank, 527 U.S. 627 (1999) Katz v. United States, 389 U.S. 347 (1967) Miranda v. Arizona, 384 U.S. 436 (1966) Planned Parenthood v. Casey, 505 U.S. 833 (1992) Roe v. Wade,410 U.S. 113 (1973) Seminole Tribe of Florida v. Florida, 517 U.S. 44 (1996) Silverman v. United States, 365 U.S. 505 (1961) Tennessee v. Lane, 541 U.S. 509 (2004) Terry v. Ohio, 392 U.S. 1 (1968) United States v. Lopez, 514 U.S. 549 (1995) United States v. Ramsey, 431 U.S. 606 (1977) Wickard v. Filburn,317 U.S. 111 (1942) Scherer, Miller / The Federalist Society’s Influence 13