Preaching from the White House: The Influence of Political Party Affiliation on Religious Signaling in National Presidential Addresses Elizabeth Chambers Undergraduate Student Department of Communication and Journalism & Honors College University of Wyoming August 1, 2019 2 Abstract This study investigated the influence of political party affiliation on religious signaling in American presidential speech. The study took a census of the most recent 20 years of State of the Union addresses (1999 to 2018) and Inaugural addresses from the latest five presidents from each party. It then analyzed counts and frequencies of explicit and implicit religious words to compare the amount of religious signaling by political party. Results indicated Republican presidents use a greater amount of explicit religious signaling in State of the Union and Inaugural addresses than Democratic presidents. Republican presidents also use a greater amount of implicit religious signaling in Inaugural addresses; in these speeches there was no significant difference in implicit religious signaling by party. Implications of these results to civil religion, Burke’s theory of identification, and the schism between the religious-aligned Republican Party and unaffiliated/non-Christian aligned Democratic Party were discussed. 3 Introduction Although founded on the principle of “separation of church and state,” the United States is culturally and politically Judeo-Christian, a fact reflected in various aspects of the political process, from the campaign trail to office. Relics of the Red Scare era, when Christian symbology was injected into political and public life, from our currency to our pledge of allegiance, remain a part of twenty-first century government. Religious affiliation continues to influence political behavior and perceptions, with Republicans aligned with Christianity and Democrats with the religiously unaffiliated and non-Christian. However, Christian religious affiliation and belief in God has seen an unprecedented decline in the American population. The role of religion in American life and politics is once again changing. That raises the question, why do politicians continue to use Christian religious rhetoric when speaking to an increasingly non-religious and non-Christian audience? Is it constitutional for figures like the president, who represents the country as a whole and not only specific religious constituencies, to convey religiosity from their public platform? Are there differences in religious rhetoric use along party lines? And if so, does the use of religious rhetoric perpetuate partisan alienation and animosity? Ample literature has examined the relationship between religious affiliation and political behavior and beliefs. Past studies (McKinney, 1944; Knoke, 1974; Campbell, Green, & Layman, 2011) have conclusively demonstrated how religious affiliation has influenced party affiliation and voting behavior for the past one hundred years. The literature examining the role of religion in political communication is much more limited, however, and not current enough to include presidents Donald Trump and Barack Obama. Existing studies (Coe & Domke, 2007; Kradel, 2004) analyze differences in religious rhetoric use among specific presidents or over time 4 without devoting much attention to partisan trends. Few if any studies have undertaken to analyze the difference in amount of religious rhetoric specifically based on party affiliation. At the same time, explicitly referencing God or using implicitly religious words to signal adherence to a particular religious worldview, a strategy termed religious signaling, has become increasingly pervasive in American politics (Coe & Domke, 2007). The purpose of this study, then, is to investigate if there is a difference in the amount of explicit and implicit religious signaling among modern presidents on the basis of party affiliation. At a time when identity-based politics dominate national discourse, with hyper-partisan lines drawn parallel to religious fault lines, this area of study is increasingly relevant and important. The historic first of Barack Obama’s presidency and the unprecedented nature of Donald Trump’s also make it imperative that literature on the “modern” presidency is updated; conclusions made prior to the last ten years may no longer be valid or hold the same relevance. Additionally, the demographic trend of secularization and religious diversification continues alongside an increase in religious signaling by politicians. The paradox requires additional study of the relationship between religion and politics in America today. This study will review literature that illuminates the historical relationship between religion and politics as well the current interplay of the two in the United States today. A complete understanding of the concept of separation of church and state requires knowledge of a time before the notion existed, when all human societies were theocracies. After an overview of theocracy, this study will outline Enlightenment-era concepts of rational and contractual governance that underpinned the first governments founded on a separation of church and state, France and the newly-established United States. It will then examine the history of practicing this principle in the United States as well as the evolving role of religious affiliation in American 5 politics. Finally, it will review the literature on the emergence of civil religion and the rise of religious signaling. Literature Review A Brief History of Theocracy Controversy over the separation of religion and politics is a distinctly modern phenomenon given most early civilizations were ruled theocratically, with no distinction between religion and government. In a theocracy, those who acted with religious or political authority were one and the same, and the state acted in the name of religious authorities such as gods (Darity, 2008). Before the development of the sociopolitical state, the power of leaders in chiefdoms was legitimated by supernatural beliefs (Scupin, 2012). Theocratic governance was evident in larger societies including ancient Tibet, where kings represented the will of deities, and ancient Egypt, where pharaohs were considered divine or semidivine and executed their earthly governance through priests. This deification of rulers also occurred in the Americas in ancient cultures including the Aztecs, Mayas, Toltecs, and Natchez (Darity, 2008). Calling on God or gods was the basis for every act of governance. Though theocracy has included as many religions as there are states, Judeo-Christian religion has played a central part in shaping modern notions of religion and politics. The concept of “theocracy” as a distinct form of government was first used to describe the first-century Jewish state, in which the law of scripture was also the law of Israel and Moses ruled in the name of God (Darity, 2008). The idea of the “secular” and a separation of religious and political authorities can be traced to the advent of Christianity, in a biblical passage where Jesus separates the realms of Caesar and God (Benz & Lindberg, 2014). Since its founding, the relationship 6 between the Christian church and state governments has had a diverse history. Before Christianity was the official religion of the Roman Empire, its followers preached respect for secular power so long as it did not interfere with divine law; emphasis on the supremacy of divine law increased when Christianity became the religion of the empire (Benz & Lindberg, 2014). The medieval era was characterized by Catholic Church-sanctioned state rulers and continual power struggles between papal authorities and kings or emperors. In the sixteenth century, the Protestant Reformation and subsequent splintering of Christianity into various sects further diminished the power of the Church relative to secular rulers. Though theocracy and Christian-aligned governance would outlast the Reformation, the time was ripe for a paradigm shift. Enlightenment Ideals A European intellectual movement of the seventeenth and eighteenth centuries, the Enlightenment celebrated human reason. Further, it advocated for a morality of reason and the use of reason to expand our understanding of the universe and to improve the human condition. Acceptance of the heliocentric model of the universe and Newtonian physics upended Christianity’s monopoly on explaining the nature of the world (Treasure, 2019). This was a blow to the Christian faith, which had laid claim as the sole authority on not only morality but also history and the natural world. To a certain extent, the Enlightenment represented a rejection of Christianity in favor of the Classical, such as the artistic achievement of the Greeks and the civil law practiced by the Romans (Treasure, 2019). The movement also embraced a quintessentially “new” way of thinking; French philosopher Voltaire criticized “the ages of belief” as backward 7 and looked to the principles of the Scientific Revolution, such as empiricism and experimentation, to guide society (Treasure, 2019). Prior to the seventeenth century, few believed a society could remain unified and stable while allowing the practice of multiple religions separate from one sanctioned by the political establishment (Benz & Lindberg, 2014). However, the religious persecution of various Protestant minorities led to a shift in opinion toward tolerance. When France’s Louis XIV revoked a policy granting religious liberty, causing the flight of 200,000 Huguenots, English philosopher John Locke penned an essay on tolerance that later informed his political ideas about the sovereignty of the people and a contract between the ruler and the ruled (Treasure, 2019); only oppressive and authoritarian governments tried to control the personal beliefs of their citizens, he concluded. Rationalism and the idea that God was subservient to reason also informed visions for the ideal government. At the core of these utopias was natural law, the idea that all people have inalienable rights like the freedom from coercion and liberty to practice their religion. These ideas provided the intellectual foundation for the French and American revolutions, which produced the first governments ideologically committed to the separation of church and state. Religion & Politics in America History. The earliest settlements in what would be the United States were ostensibly founded by Puritans fleeing religious persecution in England in order to safeguard religious liberties. In practice, however, they established theocratic colonies that only safeguarded the liberty to practice their own religion and rigidly excluded all others, including Catholics, Quakers, Jews, and native tribes with their own religions (Davis, 2010). Prior to the creation of 8 the Constitution, in various states the sanctions against these other religious groups included being barred from public office, banishment, or even execution (Davis, 2010). Public opinion on the relationship between religion and politics began to shift when future president and Constitution author James Madison reframed the issue: any government sanction of religion was a threat to religion in general because of the specific faiths it necessarily excluded. Curtailing government establishment of religion should therefore maximize the liberty to practice religion individually (Benz & Lindberg, 2014). While some Protestants stood to lose positions of power should government be separated from religion, other religions stood to gain. American Baptists, for example, celebrated the separation of church and state as a bulwark for the own faith and included the idea as a principle in their creed (Benz & Lindberg, 2014). In an essay arguing against a proposal of state-supported Christian teaching, Madison also called on the Enlightenment concept of natural rights including freedom from coercion and the right to practice one’s own religion. “The Religion then of every man must be left to the conviction and conscience of every man,” he declared, “and it is the right of every man to exercise it as these may dictate. This right is in its nature an unalienable right” (Madison, 1785, p. 1). After winning a revolutionary war against England, where the head of state and head of the church were one and the same, and emerging from a bloody history of religious conflicts prior to its settlement, the United States was established as a secular republic. This commitment is affirmed in our founding document, the Constitution. As Davis (2010) observes, the document is notably devoid of any references to God, except the pro forma “year of our Lord” date. Article VI states elected and appointed officials “shall be bound by Oath or Affirmation, to support this Constitution; but no religious Test shall ever be required as a Qualification to any Office or public Trust under the United States” (U.S. Const. art. VI). The first amendment made to the 9 document begins, “Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof” (U.S. Const. amend. I). These grand political ideas were not always reflected in the behavior of the Protestant majority toward religious minorities. From the early nineteenth century and continuing until the twentieth century, members of the majority spewed hateful rhetoric and enacted violence against Catholics, Jews, and members of a newly-founded religion, Mormonism (Davis, 2010). At the same time, however, the government held to the principle of a secular republic. Ironically, in the 1840s this diligence was fueled by underlying religious prejudice—fear that any government funds would go toward church activities after a recent wave of Roman Catholic immigration (Benz & Lindberg, 2014). In the early twentieth century, the courts enforced a strict interpretation of the First and Fourteenth Amendments in relation to religion in schools, such as the Engel v. Vitale decision that prayer in public schools was unconstitutional, even if students had the choice to abstain (Benz & Lindberg, 2014). This secular orientation was challenged by a new threat in the 1940s and 1950s: the fear of atheism. During the post-World War II Red Scare era and rise of McCarthyism, the question of religious affiliation became entwined with political creed, i.e. “godless” communism. Classical Marxism viewed religion as repressive means for maintaining the economic status quo; totalitarian regimes like the Soviet Union operating under the auspices of Marx’s idea of a communist society used this to ban the practice of religion and demand absolute loyalty to the state. In response to the threat of secularization in the U.S., elements of the Judeo-Christian tradition were injected into American political life, for example adding the new motto “In God We Trust” to the national currency and one nation “under God” to the pledge of allegiance (Greenberg, 2002). A “new religiosity overran Washington,” evident in President Eisenhower’s 10 introduction of prayer breakfasts, the creation of a congressional prayer room, and the introduction of Constitutional amendments requiring Americans to obey the law of Jesus Christ (Green, 2002, para. 8). Proving religiosity was synonymous to proving loyalty to one’s country, and those who failed to do so were attacked for supposedly having communist sympathies. The political consequences of personal religious orientation are evident in some of the earliest research done on the role of religion in American politics. A World War II-era study asked whether religious affiliation was a boon or handicap for Congressional candidates (McKinney, 1944). The analysis of election results from 1935 to 1943 found 80% of elected officials had the same affiliation as their predecessor and in fact religious affiliation was more important to predicting success than party affiliation. While the sociopolitical fabric of the country has changed in the intervening three quarters of a century since this study was conducted, it demonstrates an early interest in the role of religion in politics. The importance of religious affiliation in politics only grew during the 1950s, and arguably has not lost its salience for politicians even while its importance has waned for the everyday American citizen, as demonstrated by recent demographic changes. Today. Religious affiliation in America has undergone a dramatic shift in recent decades. Pew Research Center (2014) found in a national survey five years ago that 70.6% of Americans identify as Christian, 22.8% as unaffiliated, and 5.9% as belonging to non-Christian religions. From just seven years before, the Christian share of the population had declined nearly 8% while the ranks of the unaffiliated increased 6.7% from 16.1% in 2007. In general, the number of people identifying as Evangelical Protestant, Mainline Protestant, and Catholic has decreased while the unaffiliated and non-Christian faiths have grown. This marks an acceleration of a century-long trend of secularization and diversification away from Christian faith. In 1948 11 Gallop found just 2% of Americans identified with no religion, a figure that doubled by the 1960s and again by the turn of the century; the share of non-Protestant and non-Catholic religious adherents more than doubled from 4% in 1948 to 9% in 2008 (Newport, 2009). Belief in God, though consistently above levels of religious affiliation, has also decreased from 97% in 1965 to 86% in 2014, down 11% in less than 50 years (Johnson, 2016). While the religious landscape of the United States has changed in the last half century, the legacy of the “religious reawakening” of the 1950s is still evident in more than our national currency and Pledge of Allegiance. Even after the Supreme Court had ruled it unconstitutional, a Massachusetts mandatory-pledge bill became a flashpoint in the 1988 presidential race; George Bush attacked Democratic nominee Michael Dukakis for vetoing it while he was governor of the state (Greenberg, 2002). In 1960 Democratic presidential candidate and Catholic John F. Kennedy made a major speech declaring his allegiance to America over the Pope; in 2008, Republican contender Mitt Romney addressed parallel concerns about his allegiance to the Church of Jesus Christ of Latter-day Saints (Davis, 2010). Religious views have informed government policy domestically and abroad. For example, President George W. Bush’s 2003 President’s Emergency Plan for AIDS Relief contained the controversial requirement that one third of prevention spending be used for promoting abstinence until marriage, despite evidence this strategy is not effective for reducing infection rates of HIV/AIDS (Bendavid, 2016). The number of lawsuits involving discrimination related to religion is growing faster than those involving race or sex, and the question of separation of church and state has played out in the courts over the last several years with cases challenging the requirement for employers to provide contraception in employee health insurance plans, for federally-funded adoption agencies to consider unmarried or same sex couples, for a baker to provide his services to a gay 12 couple or transgender individual, and a slew of others (“Separation of church and cubicle,” 2015). The consequences of government sanctioning religious beliefs are very real and affect a variety of aspects of everyday life. In addition to the real-life consequences of blurring the line between church and state, the symbolic effects can be damaging. President George W. Bush spurred a trend of media coverage and concerns over the amount of religious rhetoric he used both on the campaign trail and in office (Kradel, 2004). His response to the tragedy of September 11 framed the events in highly religious terms, calling U.S. action against terrorism a “crusade” and “a battle of good versus evil” (qtd. in Steyn, 2007). That categorical hatred fueled widespread islamophobia that continues to this day, evident in politics with President Donald Trump’s 2017 ban on travel from Muslim-majority countries. A president’s symbolic preference for Christianity also calls into question whether that preference is only symbolic or if it informs preferential treatment of certain segments of society. As Steyn writes, “When a presidential regime's Christian ritual ethos accompanies output that is skewed in this manner, a society diverse with many religious and secular vocations has a right and a duty to be suspicious that governance is not being exercised fairly, that it is failing to respect the diversity of religious expression that is expected in a democracy” (Steyn, 2007, para. 35). Religious rhetoric has real-life implications. Past studies (Knoke, 1974; Coe & Domke, 2007; Campbell, Green, & Layman, 2011, Pew Research Center, 2014) have undertaken to discover the extent that religion plays a role in American politics, both demographically and symbolically. One study (Knoke, 1974) found a significant association between church attendance and party identification; five national surveys conducted over 16 years showed higher rates of church attendance were related to non- Democratic preferences for Protestants and Democratic preferences for Catholics. “Excepting 13 perhaps only race and social stratification…no other variable accounts for as much variation in party identification and voting behavior as religious membership” (Knoke, 1974, p. 51). Coe and Domke (2007) confirmed more recently that religious affiliation and behavior has become one of the strongest predictors of presidential voting patterns. Current research (Campbell, Green, & Layman, 2011) finds Evangelical Christians as a group are viewed as Republican while Catholics lack a partisan association. Identifying a political candidate as Evangelical increases Republican support and Democratic opposition. In addition, Campbell, Green, and Layman’s analysis of the 2006 Congressional vote found 62.8% of those identifying as religious voted Republican while 65.2% of those identifying as nonreligious voted Democratic. In line with this research, Pew Research Center’s 2014 national survey found self-reported religious differences exist today on the basis of party affiliation or leanings: 56% of Evangelical Protestants, 44% of Mainline Protestants, and 70% of Mormons identified with or leaned toward the Republican Party, while 54% of the unaffiliated, 64% of Jews, 62% of Muslims, and a majority of Hindus and Buddhists identified with or leaned toward the Democratic Party. Religion clearly influences political behavior such as voting and party affiliation; the next section will examine how it influences symbolic communication in politics. Political Communication & Religion America’s Judeo-Christian hegemony is evident in government activities and messages like the Red Scare-era motto “In God We Trust” and current reflexive speech closer “God bless America.” This religious symbology does not refer to a specific Christian denomination, however, something noted by Bellah (1967) in his landmark essay “Civil Religion in America.” Bellah coined the term “civil religion” to describe the role of religion in American political 14 identity, postulating that Christian symbology and ceremony in political life had grown into its own institution, independent of churches of specific denominations. Political speeches are a primary outlet of expression for Bellah’s “civil religion,” serving as a platform for signaling a politician’s religious inclinations and agenda to constituents. Of different types of political speech, Rottinhaus (2009) found nationally televised addresses are one of the most effective ways for a president to sway public opinion, though this efficacy may be declining for modern presidents with the proliferation of media alternatives. In addition to representing the individual politician’s views and agenda, national presidential addresses are particularly important since they in effect also represent the views of the president’s party. Further, coming from the head of the executive branch and figurehead of the nation, they send messages about the agenda and values of the entire country. This is particularly true of ritualized speeches; “the ceremonial nature of Inaugural and State of the Union addresses means that any invocation of God in these contexts necessarily fuses a religious outlook with the nation’s sense of itself” (Coe & Domke, 2007, p. 64). The State of the Union, mandated by the Constitution, is essentially an update on the country and policy recommendations that until 1913 was given by the president to Congress as a written report. With the Woodrow Wilson administration, the State of the Union became instead communication between the president and the American people, delivered orally, which cemented its ritual function in American political life (Peters & Woolley, 2019). State of the Union addresses are given before a joint session of Congress at the beginning of the year, typically January. Each president in the past 20 years chose not to give a State of the Union address in the final days before his term ended. In these years, the newly sworn-in president gave a speech before a joint session of Congress with the function though not title of State of the 15 Union address. “The impact of such a speech on public, media, and congressional perceptions of presidential leadership and power should be the same as if the address was an official State of the Union” (Peters & Woolley, 2019, para. 11); given this equivalency, non-official State of the Union addresses given in years during a transfer of power were included in this analysis and referred to indiscriminately as State of the Union addresses. Like State of the Union addresses, Inaugural addresses are a consistent and ritualistic occasion and are televised nationally to a wide audience. Marking the occasion president-elects assume their post, Inaugural addresses provide an opportunity for new presidents to declare to the public their intentions with their presidency as well as signal their personal ideology (University of Washington Discourse Analysis Group, 1990). These addresses are also the new president’s first speech while holding the office and not on the campaign trail, marking a shift in purpose from speaking to specific groups of potential voters to addressing the nation as a whole, now as a public figure (UWDAG, 1990). Though Inaugural addresses are aimed at a national audience, they celebrate the victory of candidates and their parties, making party identification more salient (Coe & Domke, 2007). Thus, these speeches may focus more on a candidate’s particular platform and values, compared to State of the Union addresses, which are focused on the country rather than the president, making them rich for comparison on the basis of party identification. While State of the Union addresses allow analysis of each included president’s speech over the span of years, capturing their individual presidential speaking style, first inaugurations are a one-time event, allowing comparison between presidents over a longer time frame. Existing research on religious rhetoric in ritualized presidential speech such as the State of the Union or Inaugural addresses often manipulates time as the antecedent. Kradel (2004) took 16 a comparative view of nationally televised presidential speeches to answer whether the amount of George W. Bush’s religious rhetoric marked a significant increase from other modern presidents, defined as Reagan-era and later. He used three dichotomy/trichotomy categories to describe presidents’ religious rhetoric: universal principles vs. specific theological constructions, other-worldliness vs. this-worldliness, and sin as the absence of good vs. good and evil vs. universal sin. Some points of interest in Kradel’s findings are that George H. W. Bush used more religious language than Democrats but less than other Republicans, Democrat Bill Clinton did not use a large amount of religious language but did use six of seven categories, and Republican George W. Bush did not use the largest amount of religious language but did use all seven categories. Post-September 11, Bush used significantly more language about good vs. evil than other modern presidents. Kradel also concluded that “party identification is only slightly important for predicting the amount of religious rhetoric used by presidents” but that there were significant differences in the categories of religious language favored by each party (2004, p. 15). Within their total amount of rhetoric, Republicans chose twice as many specific religious constructions as Democrats and used more other-worldly language. Concepts of sin were used by Republicans but almost entirely absent from Democratic speech. References to evil tended to spike after acts of violence, such as the Columbine school shooting and September 11 attacks, but these historical events did not account for the long-term trend of a difference between parties in references to evil. Similar to Kradel’s comparison of presidential speech over time, Coe and Domke (2007) analyzed presidential addresses to see how religious rhetoric has changed between 1933 and 2006. They categorized religious rhetoric as “God-talk,” explicitly referring to a deity, and “faith-talk,” using religiously connotated words like “blessed” or “sacred.” Coe and Domke 17 found that the frequency and amount of religious signaling in presidential speech has doubled post-Reagan, with members of both parties using it to appeal to religiously inclined voters, particularly Evangelicals. They termed this trend of using religious signaling to woo voters the “God strategy.” A candidate’s personal beliefs or practices are irrelevant so long as he or she chooses to use religious language to signal allegiance to a particular religious group or worldview, a strategy they term religious signaling. The concept is reminiscent of Kenneth Burke’s (1969) theory of identification, that speakers use linguistic strategies to give “signs” to listeners about their similarities or shared loyalties. The more similar listeners perceive themselves to be to the speaker, the more persuasive the speaker is, which explains politicians’ enthusiasm in pursuing the vote of Evangelicals and other Christians through extensive religious signaling. The increase in use of the God strategy, Coe and Domke argue, demonstrates that a politician’s religion has taken on an unprecedented importance on the campaign trail even while overall religious inclinations may be declining. “Religion has always been a political subtext in the United States, but in the past few decades the strategic use of religion has become ascendant in a manner not seen in modern American political history” (Coe & Domke, 2007, p. 2). They note that while previously Bellah’s civil religion and its expression in political speech was viewed as benign and nonpartisan, in the modern era that has profoundly changed. The fact that the pervasiveness and strength of religious affiliation has declined while use of religious language has increased demonstrates that this use is a strategy for political gain rather than an expression of sincerely held or nationally representative beliefs. As with Burke’s theory of identification, listeners need only perceive their similarity to the speaker for it to influence their receptivity to persuasion. 18 In sum, religion has long been a part of politics, both in recent history in the United States and more broadly throughout human history around the world. Early societies, as well as some nations today, were theocracies, relying on the invocation of the divine to justify their political rule. Enlightenment-era ideals of rational governance and a contract between the rulers and the ruled overturned the ascendancy of theocracy in favor of democracy with a separation of religion and government. This separation became a founding principle of the United States, despite the dominance of Protestants, members of the majority religion, in public and political life. Post-World War II, the threat of atheism eroded the separation of church and state and renewed the importance of politicians’ religious affiliation. Non-denominational Judeo-Christian symbology remains a part of American politics today, a phenomenon termed civil religion. The modern influence of religion on politics is also evident in research on the differing voting patterns, party affiliation, and public perception of political parties’ religious leanings, which overall indicates Republicans are aligned with Christianity and Democrats with the religiously unaffiliated and non-Christian. Nationally televised political speeches such as the State of the Union and Inaugural address are ritualistic occasions that convey civil religion and provide presidents with an opportunity to demonstrate their ideology to the public. Past research has shown that presidents of both parties, but Republicans in particular, are increasingly using explicit references to God and implicit Judeo-Christian associated words in these speeches to convey their allegiance to the Christian worldview, a behavior termed religious signaling. The purpose of this study, then, is to investigate if there is a difference in the amount of explicit and implicit religious signaling in the presidential speech of modern Republicans and Democrats. With this purpose, the following hypotheses were derived: 19 H1: Republican State of the Union addresses include more explicit religious signaling than Democratic State of the Union addresses. H2: Republican State of the Union addresses include more implicit religious signaling than Democratic State of the Union addresses. H3: Republican Inaugural addresses include more explicit religious signaling than Democratic Inaugural addresses. H4: Republican Inaugural addresses include more implicit religious signaling than Democratic Inaugural addresses. Method Texts The sampling unit of analysis was each speech, annual State of the Union address in the first part of the study and first Inaugural address in the second part. The first sample consisted of a census of the most recent 20 years (1999 to 2018) of nationally televised State of the Union addresses. Given the variability in speaking style and party platform adherence between presidents even of the same party, a random sample of addresses over a larger time frame would likely not provide a representative comparison between parties. The 20-year time frame falls within the period of the Reagan-era and later “modern presidency” and includes a balance of addresses given by Democratically affiliated presidents (n = 10) and Republican affiliated presidents (n = 10). The census includes two presidents from each party, with eight speeches from Republican George W. Bush (2001 to 2008), two speeches from Republican Donald Trump (2017, 2018), eight speeches from Democrat Barack Obama (2009 to 2016), and two speeches from Democrat Bill Clinton (1999, 2000). Mean speech length was 6,048 words; mean 20 Republican speech length was 5,051 words and mean Democratic speech length was 7,045 words. Transcripts of the addresses were retrieved from the American Presidency Project website (www.presidency.ucsb.edu/sou.php). The second sample consisted of a census of the most recent five Inaugural addresses from each party. Using Inaugural addresses from 1961 to 2017 places this analysis outside the bounds of the “modern” presidency but serves to capture a greater variety of speakers from each party. For Republicans, this sample included Donald Trump (2017), George W. Bush (2001), George H. W. Bush (1989), Ronald Reagan (1981), and Richard Nixon (1969). For Democrats, it included Barack Obama (2009), Bill Clinton (1993), Jimmy Carter (1977), Lyndon B. Johnson (1965), and John F. Kennedy (1961). Only each president’s first Inaugural address was used, if there was a second. Republican Gerald Ford, who succeeded following Nixon’s resignation and gave no Inaugural address, was excluded. Mean speech length was 1,800 words; mean Republican speech length was 1,980 words and mean Democratic speech length was 1,619 words. Transcripts of the addresses were retrieved from the American Presidency Project website (https://www.presidency.ucsb.edu/documents/presidential-documents-archive- guidebook/Inaugural-addresses). Procedure To measure the amount of religious signaling, syntactical context units of explicit and implicit religious words associated with Judeo-Christian tradition were coded and counted in each address. Given the possible significant difference in mean speech length, frequency of religious words based on total speech length was also calculated. Counts and frequencies of explicit and implicit religious words were averaged and then compared by party affiliation. The 21 author of this study was the only coder of the speeches. With one coder, it was not possible to asses intercoder reliability, however, to ensure coding consistency each speech was coded twice using predetermined and exclusive lists of explicit and implicit religious words. Instruments Explicit religious words were defined as Coe and Domke’s “God talk,” or making “direct reference, often by name, to a supreme being” (2007, p. 60). A list of implicit religious words (see appendix, p. 32) was adapted from Coe and Domke’s list of “faith talk,” or other religious terminology excluding direct references to a deity. Terminology exclusively related to non- Judeo-Christian religions (i.e. Buddhist, Islamic, and Muslim) was removed from the list to focus on religious signaling as an expression of America’s civil religion. Though many of the included implicit religious words are used by non- Judeo-Christian religions or in a secular sense, they are included here because of their meaning to American audiences in the context of civil religion. Thus, the coding categories should have adequate content validity. Results Mean count and frequency of explicit and implicit religious signaling in State of the Union and Inaugural addresses by party are shown in Table 1 (p. 22). The hypotheses were analyzed using independent t tests. Hypothesis 1 proposed that Republican State of the Union addresses include more explicit religious signaling than Democratic State of the Union addresses. The results indicate the mean Republican count is not significantly higher than the mean Democratic count of explicit 22 religious words at the .05 level (t = .89, df = 18, p = .19). Equal variances were assumed. On the basis of count, H1 is not supported. However, in this case word count may have less validity as an operational definition for amount of religious signaling than word frequency because there is a significant difference in mean State of the Union speech length at the .05 level (t = -6.20, df = 18, p = .00). Democratic speeches are longer. Given this, frequency of religious signaling was analyzed by comparing the ratio of religious word number divided by total speech length in words. The results indicate the mean Republican frequency is significantly higher than the mean Democratic frequency of explicit religious words at the .05 level (t = 2.33, df = 18, p = .016). Thus, on the basis of frequency H1 is supported. Table 1 Mean Comparison of Religious Signaling in National Presidential Addresses State of the Union Inaugural Republican Democratic Republican Democratic n 10 10 5 5 Speech length 5,051 7,045 1,980 1,619 Explicit count 2.60 (1.26) 2.20 (0.63) 5.20 (1.79) 3.40 (1.14) Implicit count 18.10 (4.70) 14.20 (4.10) 14.40 (4.51) 14.20 (4.66) Explicit % 0.051 (0.023) 0.033 (0.0094) 0.28 (0.13) 0.21 (0.050) Implicit % 0.36 (0.093) 0.20 (0.059) 0.75 (0.29) 0.90 (0.29) Note. Mean counts and frequencies are followed by standard deviation in parenthesis. Explicit count = number of explicit religious words per speech, explicit % = number of explicit religious words divided by total speech length, etc. Speech length is rounded to the nearest word; counts and frequencies are rounded to two significant figures. 23 Hypothesis 2 proposed Republican State of the Union addresses include more implicit religious signaling than Democratic State of the Union addresses. Using word count, the results indicate the mean Republican amount of implicit religious words is significantly higher than the mean Democratic amount at the .05 level (t = 1.98, df = 18, p = .032). Using word frequency, the same results were found; mean Republican amount of implicit religious words is significantly higher than the mean Democratic amount at the .05 level (t = 4.53, df = 18, p = .00). Equal variances were assumed. Thus, H2 is supported. Hypothesis 3 proposed Republican Inaugural addresses include more explicit religious signaling than Democratic Inaugural addresses. The results indicate the mean Republican count is significantly higher than the mean Democratic count of explicit religious words at the .05 level (t = 1.90, df = 8, p = .047). Equal variances were assumed. There is no significant difference in mean Inaugural address speech length at the .05 level (t = 1.27, df = 8, p = .12), so frequency of religious signaling was not used. On the basis of count, H3 is supported. Hypothesis 4 proposed Republican Inaugural addresses include more implicit religious signaling than Democratic Inaugural addresses. The results indicate the mean Republican count is not significantly higher than the mean Democratic count of implicit religious words at the .05 level (t = .069, df = 8, p = .47). Equal variances were assumed. Thus, H4 is not supported. Discussion This study investigated the influence of political party affiliation on religious signaling in national presidential addresses in the U.S. With three out of four hypotheses supported, overall the results showed there is a link between Republican party affiliation and increased religious signaling. This is consistent with evidence of differences in religious affiliation and religiosity by 24 political party (Knoke, 1974; Campbell, Green, & Layman, 2011, Pew Research Center, 2014). Specifically, in State of the Union addresses in the last 20 years results indicate Republican presidents use explicit and implicit religious words at a greater frequency than Democratic presidents. In Inaugural addresses for the last five presidents from each party, results indicate Republican presidents use a greater number of explicit religious words. The difference between Republican and Democratic implicit religious word use was not statistically significant. (Because mean speech length was significantly different by party for State of the Union addresses, these speeches were compared on the basis of word frequency while Inaugural addresses, which had no significance in length by party, were compared on the basis of word count.) Based on these results, the temptation is to classify Republicans as using large amounts of religious rhetoric and Democrats as not. However, within this overall trend some measures showed religious rhetoric use as approximately equal between both parties, indicating a greater tendency of politicians to use religious rhetoric. For example, results showed no significant difference in the number of times Republican and Democratic presidents reference God in State of the Union addresses. The fact that there are consistent levels of referencing a Judeo-Christian deity (i.e. explicit religious words) in this annual ritualistic presidential speech supports Bellah’s (1967) notion of civil religion. In Inaugural addresses, results also showed there was no significant difference in the use of implicit religious words with Judeo-Christian connotations, further supporting the idea of Christian symbology and ritual embedded in American politics. This consistent minimum level of God-talk supports Coe and Domke’s finding of high levels of religious signaling in the modern era. Results found there is a significant difference in the amount Republican and Democratic presidents used explicit and implicit Judeo-Christian words in State of the Union addresses and 25 the explicit words in Inaugural addresses. This indication that Republicans generally do more religious signaling than Democrats adds a layer of complexity to previous findings (Coe & Domke, 2007) that both modern Republicans and Democrats use religious signaling extensively. However, these results do not break entirely from previous research since they found no Democratic president completely avoided religious signaling. By analyzing speeches up to 2018, this study updates existing literature on religious signaling in presidential speech and may reveal new trends since Coe and Domke’s analysis. Since few studies have analyzed the amount of religious signaling on the basis of party affiliation rather than change over time, this study also contributes more generally to the body of knowledge on the relationship between religion and politics in the U.S. Significance The intersection of religion and politics in the U.S. is an important area of study in an era of increasingly polarized politics. Pew Research Center’s 2014 report indicates nearly a third of Americans do not identify as Christian, with growing segments identifying as unaffiliated or non-Christian. This raises questions about the impact of religious signaling in political speech, especially presidential speech. Given a widening demographic divide on religion, does religious signaling in political speech serve to unify or divide the populace? It may be point of solidarity for Christians and a cause for alienation for unaffiliated and other religious groups. Religion embedded in government has become a flashpoint for controversy and ire, as evidenced by increasing rates of litigation concerning religion and by a number of high-profile Supreme Court cases. So, why do presidents continue to use religious signaling? The popularity of Coe and Domke’s God strategy must be due to its effectiveness. It makes flagrantly irreligious 26 politicians like Donald Trump appealing to Christian voters, in keeping with Burke’s (1969) notion of linguistic signs of similarity increasing persuasion. Religious signaling may be a token nod to Christian voters in the interests of maintaining popularity and approval, or it may more insidiously indicate preferential treatment for this segment of the population. Whether symbolic or actual, this preference threatens the founding Constitutional promise to make no establishment of religion and in effect exclude all unsanctioned others. Given the history of violent religious conflict on American soil and around the world, maintaining religious neutrality and tolerance in government is critical. Putting personal identity and alleged or real religious group allegiance before the interests of all citizens, regardless of their religious views, is an abuse of power and an affront to Enlightenment ideals of rational governance and respect for natural rights. Although a Congressional, state or local politician representing a highly religious constituency would be justified in promoting those interests and worldview, the president has a duty to represent the interests of the entire American people. That people is not the bastion of Christianity and belief in God that it once was. Limitations & Future Research There are several limitations in this study. Though it attempted to balance breadth and depth of material analyzed by considering 20 years of annual State of the Union addresses and 56 years of one-time Inaugural addresses, the type of text used limits the scope of the study. State of the Union and Inaugural addresses are useful for comparative purposes since they are both nationally televised and remain constant in purpose and audience, but they are not a comprehensive look at all of a president’s public communication. Amount of religious signaling likely changes when presidents speak on more specific occasions and to more specific audiences, 27 for example, to the Christian Coalition of America rather than a national audience. Only speeches aimed at a national audience were included in this study in the interests of a valid comparison, but considering more specific speeches such as these would undoubtedly result in a more comprehensive look at the extent of presidents’ religious signaling. In addition to the texts used, the instrument used to measure religious signaling is also limiting. Coding only for words from a set list limits the validity of the study to measure religious signaling, though it does help ensure consistency in coding. Coding yes/no for these words and tabulating the totals precludes the nuance of noting which words were used and in what way, as could be done in a more qualitative analysis like Kradel’s (2004). Finally, presidents vary considerably in individual ideology and speaking style, which limits the degree to which they can represent the values of the parties they belong to and the degree to which we can generalize from presidents to parties. In addition, political parties in the U.S. are fragmented by federal, state and local divisions. These two factors mean presidential speeches may be more representative of individual style rather than party orientations as a whole. There are other areas future research could investigate concerning religious signaling in political speech. Chief among these is analyzing other presidential speeches besides State of the Union and Inaugural addresses to gain a more holistic view of executive religious signaling. Future research should also collect data from other levels (state and local) and branches (legislative and judicial) of government. Anecdotally, this study found speeches with similar or identical counts of implicit religious words often used entirely different words on the list; a more detailed qualitative analysis could investigate which religious words or types of words are favored by each party, as done by Kradel. This study assumed the specific religious beliefs of each president are irrelevant since it is impossible to measure the sincerity of these beliefs, but 28 future studies could undertake to study how personal affiliation or religious involvement influences religious signaling in political speech. In sum, the intersection of religion and politics in the U.S. is a rich area of study. 29 References Bellah, R. (1967). Civil religion in America. Daedalus, Journal of the American Academy of Arts and Sciences, 96, 1-21. Bendavid, E. (2016). Promoting abstinence, fidelity for HIV prevention is ineffective. Retrieved from Standford Medicine News Center: http://med.stanford.edu/news/all- news/2016/05/promoting-abstinence-fidelity-for-hiv-prevention-is-ineffective.html Benz, E. W., & Lindberg, C. H. (2014). Christianity: Church and state. Retrieved from Encyclopædia Britannica: https://www.britannica.com/topic/Christianity/The-Christian- community-and-the-world Burke, K. (1969). A rhetoric of motives. Berkeley: University of California Press. Campbell, D., Green, J., & Layman, G. (2011). The party faithful: Partisan images, candidate religion, and the electoral impact of party identification. American Journal of Political Science, 55(1), 42-58. Coe, K., & Domke, D. (2007). Speaking to the faithful: the rise of the God strategy in American politics. Conference Papers – National Communication Association, 1-32. Darity, W. A. (2008). Theocracy. In International encyclopedia of the social sciences (2nd Edition ed., pp. 342-343). New York: Macmillan Reference USA. Davis, K. C. (2010). America’s true history of religious tolerance. Retrieved from Smithsonian.com: https://www.smithsonianmag.com/history/americas-true-history-of- religious-tolerance-61312684/ Greenberg, D. (2002). "Under God"--A product of the 1950's Red Scare. Retrieved from George Washinton University: History News Network: https://historynewsnetwork.org/article/840 30 Johnson, D. (2016). See how Americans' belief in God has changed over 70 years. Retrieved from Time: http://time.com/4283975/god-belief-religion-americans/ Knoke, D. (1974). Religious involvement and political behavior: A log-linear analysis of white Americans, 1952-1968. Sociological Quarterly, 15(1), 51. Kradel, A. (2004). God on our side: The religious rhetoric of recent U.S. presidents. Conference Papers – American Political Science Association, 1-17. Madison, J. (1785). Memorial and Remonstrance against Religious Assessments, [ca. 20 June] 1785. Retrieved from National Archives: Founders Online: https://founders.archives.gov/documents/Madison/01-08-02-0163 McKinney, M. M. (1944). Religion and elections. Public Opinion Quarterly, 8, 110-114. Newport, F. (2009). This Christmas, 78% of Americans identify as Christian. Retrieved from Gallup : https://news.gallup.com/poll/124793/this-christmas-78-americans-identify- christian.aspx Peters, G., & Woolley, J. (2019). Annual messages to Congress on the State of the Union (Washington 1790 - Trump 2019). Retrieved from The American Presidency Project: https://www.presidency.ucsb.edu/documents/presidential-documents-archive- guidebook/annual-messages-congress-the-state-the-union Pew Research Center. (2014). Religious landscape study. Retrieved from Pew Research Center: www.pewforum.org/religious-landscape-study/ Rottinghaus, B. (2009). Strategic leaders: Determining successful presidential opinion leadership tactics through public appeals. Political Communication, 26, 296-316. Scupin, R. R. (2012). Chiefdoms. In Cultural anthropology: A global perspective (8th Edition ed.). London: Pearson. 31 Separation of church and cubicle: Religion in the workplace. (2015). Retrieved from The Wharton School of the University of Pennsylvania: https://knowledge.wharton.upenn.edu/article/separation-of-church-and-cubicle-religion- in-the-modern-workplace/ Steyn, M. (2007). May god continue to bless America: Religion and politics in post-9/11 America through the lens of Buddhist philosophy. Journal for the Study of Religion, 20(2), 77-100. Treasure, G. R. (2019). History of Europe, the Enlightenment. Retrieved from Encyclopædia Britannica: https://www.britannica.com/topic/history-of-Europe/The-Enlightenment U.S. Const. amend. I. U.S. Const. art. VI. University of Washington Discourse Analysis Group. (1990). The rhetorical construction of a President. Discourse & Society, 1(2), 189-200. 32 Appendix Explicit religious signaling Instance of direct reference to a deity or supreme being: God, Christ, Lord, Creator, the Almighty, or Providence. Implicit religious signaling Instance of the following roots, words, or words derived from roots: amen, angel, angels, angelic, apostle, backslid, baptism, baptize, believer, bible, biblical, bless, cathedral, christian, church, churches, clergy, commandment, communion, confession, congregation, consecrate, covenant, creed, crusade, denomination, devotion, devout, disciple, epistle, evil, faith, fellowship, fruits, genesis, gospel, grace, hallow, heaven, holy, hymn, immortal, Jew, lamp, martyr, miracle, mission, orthodox, parable, pastor, peacemaker, penance, piety, pious, pope, pray, priest, prophe, proverb, psalm, pulpit, rabbi, reap, rebirth, reborn, redeem, redemption, religio, repent, restor, resurrect, reverend, sabbath, sacrament, sacred, saint, salvation, sanctity, sanctify, sanctuary, scriptur, sermon, servant, shrine, sin, sins, sinner, sinners, sinning, sinned, solemn, soul, sow, sows, sown, sowed, spirit, temple, testament, theolog, trinity, worship.