Who speaks for the people? The President, the press, and public opinion in the United States.

Sparrow, Bartholomew H.

The U.S. president, the press (or media), and public opinion survey data are all stand-ins--substitutes--for the American people. In the United States as a federal republic, the U.S. president is the sole elected representative of all Americans. Among the president's chief attributes is his role as a rhetorician and national communicator; one person, the president, gives voice to the mass public. He serves as the personification and the symbol of the United States (Ceasar et al. 1981; Cohen 2004; Hart 1987; Wattenberg 2004). Furthermore, the president's role as chief communicator has dominated media attention over the last several decades, albeit at the expense of Congress members, other members of government, and other members of his political party (Rozell 2003). What the public learns about government often comes from what the president imparts.

The press, for its part, provides the forum for and content of public discourse. Reportage by journalists of what politicians, government officials, businesspersons, and professionals say or write; published or spoken commentary by known figures; guest editorials; letters to the editor; and Web logs by known and unknown members of the public constitute the public sphere. Almost all of what Americans know about national politics, the U.S. government, their fellow citizens, and the larger world is communicated through the media. The persons, ideas, and arguments of national politics and government are what members of the public absorb from watching television; reading newspapers, magazines, and other publications; and being on the Internet. The press (or media, to use the terms interchangeably) may distort political reality in predictable ways (Bennett, Lawrence, and Livingston 2007; Sparrow 1999; Cook 1998), but for practical purposes media reality is political reality.

Public opinion polls, too, speak for the American public. If public opinion had once been an amalgam of public correspondence, politicians' conversations, letter-writing campaigns, petitions, and public demonstrations, this has not been the case for more than a half century. Scientific public opinion surveys have effectively made public opinion identical to polling results, and polling results are typically now the only indicator used for the determination of popular views and personal behaviors with respect to particular persons and issues. Vox Populi, Vox Dei. Dick Cheney's infamous recent response (the vice president replied "So?" to an ABC News interviewer's declaration that two-thirds of Americans believe that the war in Iraq was not worth fighting) is the exception that proves the rule (Raddatz 2008): Few politicians or public figures can publicly speak out against, or voice opposition to, the American public. And very few politicians or officials, if any, can do so consistently. On the contrary, politicians, government officials, and the public pay attention to public opinion reflected in polling data. While public opinion may not ultimately settle issues, it almost always factors in decision making, as accounts of the operations of the Bill Clinton and George W. Bush administrations indicate. And if public opinion is especially one sided, it may actually be controlling.

If this introductory set of comments seems obvious, perhaps less obvious is how these three institutions--the presidency, the press, and the polls--interrelate. They cannot all represent the public's voice, obviously, unless they are in full agreement. Yet much of the time, and possibly most of the time, this is clearly not the case. Does one of these institutions better represent the public than the other(s)? Does any one (or any two) dominate the other(s)? Or, are there separate spheres in which one (or more than one) of these public voices is (are) privileged?

The purpose of this article is to reconsider these three institutions--the first, formal, the latter two, informal--and their role in the political system. This is no trivial matter. The Preamble of the U.S. Constitution begins by declaring, "We the people of the United States...."; the United States was created by the people. In Chief Justice John Marshall's words for a unanimous Supreme Court in M'Culloch v. Maryland (17 U.S. 316 [1819]), the government of the United States is "emphatically and truly, a Government of the people. In form and in substance, it emanates from them. Its powers are granted by them, and are to be exercised directly on them, and for their benefit" (404-05). And in the Gettysburg address, President Abraham Lincoln famously proclaimed the United States to be a government of the people, by the people, [and] for the people." The matter of who speaks for the American people is of great symbolic, rhetorical, and ultimately political importance. It is especially important at a time when Americans appear to have ever less trust in the institutions of government, business, and the leading professions. Neither do people identify more with their state than the nation: Not only do voters turn out at lower levels for state and local races (or ballot referenda) than for federal races, but the state governments exercise fewer and fewer distinct powers.

Let me proceed in the order of the three institutions introduced above and then offer some concluding remarks. First, however, a caveat: By constitutional design the U.S. House of Representatives is the first branch of government and, as such, closest to the voters. Indeed, analysts of the "rhetorical presidency" are keenly aware that Congress's constitutional role has been usurped by what effectively consists of a second constitutional system (Tulis 1987; Medhurst 1996). For present purposes, however, I ignore the constitutional implications of the fact that Congress no longer appears to serve as the voice of the people. Why the institution of Congress is not the institution most directly tied to the American people, contrary to the structure set up under the U.S. Constitution, is an issue for another day.

The President as the Public's Voice

The fact that the president of the United States speaks for all Americans--as a mass public--is a commonplace, perhaps ever since President Franklin D. Roosevelt used national radio addresses in the 1930s and 1940s during the years of the Great Depression and World War II. The same could be said of Woodrow Wilson, Theodore Roosevelt, and William McKinley before FDR (if without the national radio broadcasts). While the president surely tries to court public opinion--and to appeal to other politicians and key opinion makers for the sake of getting policies enacted, nominees appointed, and being reelected, for the president to even consider "going public" stands as a testament to his claim as a national leader. As one close student of the presidency, George Edwards, remarks: "leading the public is at the core of the presidency" (Edwards 2003, 4).

Presidential leadership may be as much a matter of public willingness to accede to presidential leadership, as it is a matter of presidents and the bully pulpit. If popular opinions are typically resistant to change (Edwards 2003), it is also true that persons often have impressionable and changing opinions on many issues (often depending on question wording and question order). Persons may not have well-formed opinions, especially with respect to issues of less immediate relevance to their own lives (Fishkin 1991). Under many conditions, only modest changes are able to alter public opinion and reverse the minority and majority status of different issue positions.

One category in which the president speaks for the public--operationalized here as a job approval rating by a supermajority of 60 percent of public opinion--is with respect to foreign affairs and national security. Despite the passage of the War Powers Resolution and despite the end of the Cold War, the president's role as the international representative of the United States, head of the executive branch, and status as commander in chief allows him to exert virtual plenary authority over U.S. foreign policy. He has the military, diplomatic corps, intelligence community, customs and immigration agencies, U.S. attorney general and top lawyers, and others at his beck and call. With the president's close access to state-of-the-art information, intelligence, expertise, and advice, Americans, journalists, and most members of Congress generally defer to the president on issues of foreign policy, national security, international economics, and related issues (Lindsay 1992-1993; Ornstein and Mann 2006; Peterson 1994; Wildavsky 1966). Although the notion of the "two presidencies" does not always or uniformly obtain, research suggests that opposition party members have supported the president more in foreign policy than in domestic policy in the years from 1989 to 1998 (Schraufnagel and Shellman 2001) and since 2001. And without opposition on Capitol Hill and without opposing members of Congress offering competing accounts of foreign policy and national security, neither does the press (Bennett, Lawrence, and Livingston 2007; Entman 2004).

Only in circumstances of failure (e.g., post-1967 Vietnam, post-2004 Iraq) and only well after the U.S. government has committed to specific courses of action do others in government, the media, and large proportions of the public begin to seriously and persistently question the president and attempt to amend what become to be seen misjudgments and missteps (Mueller 2000, 2003). Usually, however, the exercise of presidential voice in foreign policy is a popular voice--as with President George H.W. Bush prior to, during, and shortly following the 1991 Gulf War; President Clinton deploying U.S. forces in the former Yugoslavia; and President George W. Bush's use of the military soon after the events of September 11, 2001.

Another category of presidential leadership is the president's status under unified government, when the president's party controls both chambers of Congress and the president thus becomes the point man for particular policy and ideology. Consider President Lyndon Johnson in 1965; President Carter in 1977; President George W. Bush in 2001 (despite his controversial and razor-thin electoral mandate) and 2005; and most obviously, FDR in 1933, 1937, 1941, and early 1945. Yet the president's window of popularity and goodwill may be short-lived (think of Carter by late 1978 or Bush by mid-2005), even if the president's party commands majorities in both chambers of Congress.

The president also speaks for the people when there is a landslide election putting him into office--that is, when he has a demonstrable mandate, whether by popular vote or by delegate count in the Electoral College. Consider President Johnson garnering 61.1% of the popular vote in 1964 or Richard Nixon's 60.7% in 1972. As for the Electoral College vote, recent landslides include Ronald Reagan's victory over Jimmy Carter in 1980 (489 to 49), Reagan's 1984 triumph over Walter Mondale (525 to 13), and George H.W. Bush's win over Michael Dukakis (426 to 111, with one West Virginia delegate voting for Lloyd Bentsen). When the voters directly or indirectly elect the president with overwhelming majorities, the president can be said to speak for the public.

In addition, apart from issues of foreign policy, unified government, and landslide elections, the president speaks for the people when he is especially popular (defined as above 60% poll approval ratings). Consider presidents George W. Bush (late 2001 through 2003), Clinton (1998 and 1999), George H.W. Bush (1991), Reagan (1981, 1984, and 1985), Carter (1977), Ford (late 1974), and Nixon (late 1972). The periods of particular popularity may coincide with the other factors above, of course, whether the timing of foreign policy actions (the two George Bushes), landslide electoral wins (Reagan, Carter, Nixon), control of both houses of Congress (Carter, George W. Bush), or other reasons (such as economic prosperity, as with Reagan in 1984 and 1985 and Clinton in 1998 and 1999). This category is obviously not mutually exclusive from the other classes of presidential leadership, though it may be.

Conversely, there are multiple conditions--all else being equal--under which the president does not speak for the American people. This is typically the case when domestic policy dominates and where there are almost always formidable opponents (Clinton and health care in 1993; Bush and Social Security in 2005), with divided government and when the president faces opposition-party control of one or both houses of Congress (Clinton in 1994-95; Bush in 2006-08), when the president is narrowly elected into office and thus has little claim of an electoral mandate (Bush in 2001 and 2005), and when half or more of the American public disapproves of the president's performance and therefore only half or less than half of the public approves. This was Nixon's fate in late 1971 and early 1972, Ford's experience in late 1974 and early 1975, Carter's lot in 1979 and 1980, Reagan's condition in 1987, George H.W. Bush's circumstance in 1992, Clinton's condition in late 1993 and 1994, and George W. Bush's predicament from March 2003 forward.

The demarcation of these categories supports Edwards's thesis that presidential rhetoric itself cannot bring about his personal popularity or policy success (Edwards 2003). Instead, presidential congruence with popular opinion depends on other factors: foreign policy or military actions, party control of both branches of government, overwhelming electoral success, and individual presidential popularity.

The Press as the Public's Voice

Implicit with the president's ability to be the voice of the people is the crucial factor of his relationship with the press (i.e., with the large established print and broadcast media firms). Presidential leadership largely depends on favorable coverage in the media--that is, on the secondary coverage offered in print and on the air, rather than on the direct communication afforded by the broadcast or publication of his speeches and press releases through the president's use of the bully pulpit (Edwards 2003; Kernell 1997). The president cannot simply go public; he is increasingly unable to secure multiple network channel coverage of an address or pronouncement, and those who do watch the president constitute ever-smaller proportions of the electorate who, in any event, typically remember little of what they see and hear (Edwards 2003; Wattenberg 2004). Rather, "the media independently choose which issues to highlight and how to frame them for their viewers and readers. As a result, the media provide powerful competition for the president in his attempts to structure the choices before the public" (Edwards 2003, 173). Note that Edwards's statement recognizes both the agenda-setting effects ("which issues to highlight") and framing effects ("how to frame them") of the media. The sections below discuss agenda setting and framing in turn.

In fact, Edwards recognizes that "the media are unlikely to adopt consistently the White House's framing of issues" (Edwards 2003, 184). For one, this means that the press may be aligned with the president and in such circumstances (presumably coinciding with the factors above) there more or less obtains a single national voice for the people. Under these conditions, when the media becomes a conduit or transmitter of the president and his administration or party, the press cannot be said to be the public's voice because the press does not exhibit any independent expression but instead merely serves as the echo or amplification of the president's, his administration's, or his party's perspective (Bennett, Lawrence, and Livingston 2007; Iyengar and McGrady 2007; Schiffer 2008). Edwards's statement also recognizes, though, that the press can stymie the president as political leader and the voice of the public; the substance and frames of press coverage may depart from the substance and frames preferred by the White House. In this latter case, when the press offers information and viewpoints contrary to those of the president, other political leaders, and government officials, the press can act as the public's voice.

The press, by setting the public agenda, acts as the public's voice by determining which (few) subjects are relevant for popular consumption (Edwards 2003; Iyengar and Kinder 1987; Iyengar and McGrady 2007; McCombs 2004). Empirical and experimental data show that what the press focuses on as its principal topics are the issues that public opinion holds as the most important. In fact, high levels of press attention can, in turn, drive political leaders to take action in both foreign policy and domestic policy. Whether with respect to crime or other domestic policy, a foreign nation, or other issue, what the press covers and focuses on is what the public believes is important. Should one or more of these issues not be on the president's agenda, then they become the popular agenda. The "media's issue agenda becomes the public's agenda" (Iyengar and McGrady 2007, 210). As Bernard Cohen remarked more than four decades ago, the press does not determine what people think, it determines what people think about.

Recent examples of agenda setting can be taken from the 2008 Democratic presidential primaries. Over the two-month period from March 1, 2008, to May 1, 2008, U.S. newspaper and wire services ran 251 stories (editorials included) that reported on Barrack Obama and his bowling. They also ran 148 news stories on Obama and the wearing of lapel flag pins. Meanwhile, a similar search of the LexisNexis database revealed that over the same two months, the newspapers and wires featured 169 stories on Obama and Iran and 101 stories on Obama and global warming/climate change (all terms within eight words of each other). The public agenda in late winter and early spring of 2008 was more consumed about Obama's bowling than it was about Obama's views on Iran or on global warming/climate change. (1)

Furthermore, consider the attention given to Hillary Clinton and Obama early in the Democratic contest, as opposed to the attention given John Edwards, Joe Biden, or Chris Dodd--this despite Edwards's strong showing in preprimary polls and in the Iowa caucuses (finishing second, with Senator Clinton close behind) and despite the fact that this was reportage on the presidential primaries rather on a standing government (in which case we might expect Bennett, Lawrence, and Livingston's "power indexing" to obtain [2007]). The media clearly sets the public agenda with respect to which candidates receive frequent and prominent attention. Yet if persons (and issues) are not covered in the press, they are effectively ignored and there is no (other) public voice.

Beyond setting the public agenda, the media influence how people think about issues; this is the media's framing capacity or what may be seen as the media's third face of power, to follow Stephen Lukes (Sparrow 2006). Said differently, the media not only constitutes the peoples's voice with respect to what they think about (Lukes's second face of power), but also shapes the terms by which they think about political issues. Framing refers to the media's capacity to identify and categorize particular political issues by virtue of the media's choices of and emphases on which aspects of issues to report. In other words, frames are the categories, definitions, implied parallels, and suggested solutions that bear on issues of politics and government (Iyengar 1991; Entman 1993). They are the means by which the media construct social reality. Because most events, most persons, and most concepts have multiple facets, the media define the public's voice by framing issues in (some) accessible and familiar ways--and not in other, perhaps equally accessible ways.

The frames by which the media present social reality exert persistent and self-perpetuating effects. As Iyengar points out, the episodic frames by which the television media present particular crimes or acts of terrorism (i.e., by presenting them as distinct and unrelated events) lead the American public to ignore the larger causes of and interrelationships among incidents of crime and political violence (Iyengar 1991). In fact, a small set of news frames dominate news coverage in the United States, among them frames such as ethnocentrism and nationalism, individualism and personalization, and partisanship (Page and Shapiro 1993; Neuman, Just, and Crigler 1992; Sparrow 1999). Note, however, that in almost all circumstances these frames can be countered by other frames that familiarize audiences with particular events or specific issues with alternative meanings. Ethnocentric and patriotic frames can be replaced by frames of universal or humanistic norms and values; framing in terms of individualism and personification can be substituted by framing in terms of societal factors and organizational actions; and framing though partisanship and contestation can be supplanted by frames that emphasize nonpartisanship (or bipartisanship) and consensus.

Consider, again, the example of the 2008 presidential primaries. The Democratic campaign was dominantly framed by the horserace between Senators Clinton and Obama--who was ahead, who was behind, and why (i.e., what tactics were being used)--at the expense of more serious coverage of Iraq, Iran, Afghanistan, the mortgage foreclosure crisis, the Middle East peace process, torture, presidential malfeasance, and inflation. The media predominantly frame politics in terms of the horserace between candidates and campaigns (Patterson 1993, 2004), rather than in terms of who gets what when and how (to paraphrase Harold Lasswell's definition of politics).

Yet as Elisabeth Noelle-Neumann (1983), Nina Eliasoph (1998), and others observe, if the people do not read, see, or hear in the media what they themselves think, believe, or intuit, then the people's unarticulated views become the proverbial fallen trees in the forest: unheard and unheeded. In the absence of alternative language, ideas, and public discourse, the media's own frames effectively become the lenses through which the public sees current events and recent history. Sometimes these frames are those offered by the White House, and in other instances they are not. What is more, these frames achieve even greater effect by virtue of the fact that they are often taken for granted--unstudied and accepted subconsciously. But they then become, by default, the public's frames.

Public Opinion Polls as the Public's Voice

Public opinion serves as a metric of presidential leadership with respect to presidential approval ratings. Presidents and their advisors use public opinion not as an absolute guide, but rather for tactical purposes, and instrumentally, for reaching particular political ends (Jacobs and Shapiro 2000). In general, political analysts conceive of public opinion as a channel or guide for policy makers, boundaries beyond which they cannot go but which also offer leeway in terms of the exact path policy makers take. Public opinion serves as a "permissive limit" for policy makers (Almond 1950; Key 1961; Sobel 2001).

With polling data as the accepted indicator of public opinion, though, dozens of polling groups are conceivably able to define public opinion--some university affiliated, others connected to nonprofit foundations, some linked to media firms, and others simply as independent for-profit consulting businesses--through the hundreds of thousands of poll questions that they ask each year (iPoll database). Yet few of the academic or nonprofit polling data reach the public, because the overwhelming portion of what the American public learns about public opinion comes from the major media polls. For all the quality of public opinion research being done by the National Opinion Research Center at Chicago, the National Election Studies, the Chicago Council of Foreign Relations, and other organizations, public opinion for practical purposes is the product of the mainstream media polling firms, the polls conducted by ABC News (ABC News/ Washington Post/Stanford University), CBS/New York Times, CNN, Fox, Los Angeles Times, NBC/Wall Street Journal, Newsweek (Princeton Survey Research Associates International/ Newsweek), Time, USA Today (Gallup/USA Today), and the Associated Press (Associate Press/Ipsos).

The consequence is that the public learns about public opinion from the media polls. The process happens in distinct stages. First, polls get commissioned, whereby the polling firms select a relatively few number of questions to ask their respondents out of the many possible questions about important and relevant topics. Then, the media publicize and disseminate only some of the findings from the hundreds of thousands of poll questions that their and other polling firms ask each year. In short, for all of the possible subjects about which there may be public opinion, the public learns of only a small portion of the possible public opinions available.

Consider as an example Americans' opinion on global warming/climate change in 2006 and 2007 (as a point of reference, Al Gore's An Inconvenient Truth came out in theaters in early 2006 and on DVD in late 2007). The public put "environment/ pollution" (Gallup) or "environment" (CBS/New York Times) as the "most important problem facing the country" an average of 1% to 3% of the time over the 24 months (Gallup asked the question on a monthly basis; neither Gallup nor CBS/New York Times has a category or code for global warming or climate change). Respondents' concerns over the environment understandably came well behind their concerns over the "Situation in Iraq/War" (Gallup), the "War in Iraq" (CBS/New York Times), the "Economy in general" (Gallup), and "Economy/Jobs" (CBS/New York Times).

But when Gallup asked respondents what they thought the most important problem would be "25 years from now," a question Gallup asked just once in the two-year period, respondents listed "Environment/Pollution" first, at 14%; another 8% of respondents answered "Lack of energy sources/energy crisis," tied with "Social Security" and "Health Care" (Gallup Poll, March 2007, Roper Center, University of Connecticut). Furthermore, when respondents were directly asked their opinions about global warming/climate change, they expressed strong opinions. More than four out of five respondents (83%) were either "very concerned" (53%) or "somewhat concerned" (29%) about "the way the world produces and uses energy is causing environmental problems including climate change" (PIPA/BBC World Service, June 2006, Roper Center, University of Connecticut). Furthermore, almost two-thirds of respondents (64%) felt that global climate change was under way and that public action was needed (NBC News/ Wall Street Journal, January 2007, Roper Center, University of Connecticut). More than half of respondents (54%) disapproved of current U.S. policy on global warming/climate change, in fact, with 37% strongly disapproving the United States' "handling of global warming or climate change" and 17% "somewhat disapproving" (PIPA/BBC World Service, December 2006, Roper Center, University of Connecticut). And over two out of three Americans (68%) thought that the Bush administration "Hasn't done enough" "to address climate change and reduce global warming" (Princeton Survey Research Associates International/Newsweek Poll, August 2007, Roper Center, University of Connecticut).

The iPoll database shows that people do care about global warming/climate change and want action taken on it, even if they viewed the war in Iraq and the economy as more immediate concerns. Yet the media firms' polls barely asked about global warming/ climate change; out of the hundreds of thousands of questions asked over the two years, ABC News (ABC News/Washington Post/Stanford University), CBS/New York Times, CNN, Fox, Los Angeles Times, NBC/Wall Street Journal, Newsweek (Princeton Survey Research Associates International/Newsweek), Time, and USA Today (Gallup/USA Today) asked a total of 108 questions on global warming/climate change. This is despite the significant poll findings noted above, despite the popularity and attention given to "An Inconvenient Truth," and despite the almost unanimous scientific evidence about the reality of climate change.

What little polling data there were about global warming/climate change, moreover, received scant attention in the press. This is to say that even given the little polling data on the topic and even with the public's apparent concern about global warming/ climate change and U.S. policy, there was very little print coverage of the topic in the nation's newspapers and magazines. A LexisNexis search for 2006 and 2007 revealed only 29 news articles published in the United States that reported on public opinion with respect to global warming/climate change. (2) The New York Times, Washington Post, and Wall Street Journal ran a total of only seven articles over the period, and the balance of stories ran in regional newspapers such as the Milwaukee Journal Sentinel (which ran a series on global warming in its "State and Regional" section), the Lewiston Tribune (Idaho), the Albuquerque Journal, and the San Diego Star Tribune. Perhaps as significantly, the news stories were almost exclusively printed on inside pages, and not even in the papers' main news sections (the New York Times and Washington Post each ran a feature story on global warming/climate change in their Sunday magazines).

The example of media polling and then reportage of public opinion on global warming/climate change is, of course, illustrative rather than definitive. But it makes for a strong image: the media firms decide what to ask in their polls, and then they decide which polling they will report--and where and how they will report on the topic (the Washington Post Sunday magazine's story of global warming, for instance, was critical of the very idea of global warming/climate change).

It turns out, in fact, that media firms treat public opinion polling results not so much as independent news (as a noteworthy finding worth publicizing no matter the source), but more as the product of their own investments (the analog to the fact that the media favor the reporting done in their own reporters rather than that of news services or wire services, notwithstanding the quality of the news). One recent study of public opinion on U.S. foreign policy and the media found that, when the New York Times reported polling data, it cited its own New York Times/CBS polls almost 4 out of 5 times (79%, n = 467), that the Washington Post reported on its own Washington Post/ABC polls 75 % of the time (n = 604), that the Wall Street Journal cited Wall Street Journal/NBC polls 85% of the time (n = 149), and that USA Today cited its own USA Today polls 86% of the time (n = 886). Nor were newsmagazines otherwise: Newsweek cited its own polls 87% of the time (n = 31), and Time referred to its own Time/CNN polls at an 87% rate (n = 151) (Sparrow and Stroud 2007, Table 4). The public opinion learned by the public via the media appears to be the product of the media firms doing the reporting. A media firm's own news values trumped the newsworthiness of (other) polling data, not to mention the public interest.

Benjamin Page and Robert Shapiro may be right with their view of a rational, more constant, and savvier public than news reports about public opinion would have us believe (1993). Page and Shapiro, Richard Sobel (2001), Steven Kull and I.M. Destler (1999), and others offer evidence that the public can and does exercise good judgment, shows common sense, and learns from its experience. But these longer-term, more careful studies of public opinion are not those that citizens learn about. Rather, Americans learn what gets reported in the media--not what gets written up in scholarly books and journals. There may be as many "publics" and "public opinions" as there are issues, but clearly not all questions are equal to the media. Meanwhile, only a small subset of the public-opinion survey results that are available get disseminated by the media--and even those few that get publicized by any one media firm are most likely to have been those polling results commissioned and collected by that same media firm.

Discussion

Who then speaks for the public? First, it is clear that public opinion does not well represent the public's views. What gets asked in poll questions, how questions are asked, what polling data gets reported, and how public opinion gets reported all constitute problematic links in the public-opinion process. Apart from answers to standard polling questions such as what is the top problem facing the country, whether citizens approve of the president's performance in office, if the country is going in the right direction, or which candidates respondents would vote for, the popular discourse on public opinion represents but a partial and contorted view of people's minds. While the lesson that it takes effort and organization to elicit public opinion is by no means novel (Edelman 1977; Fishkin 1991; Gamson 1984; Ginsberg 1986; Herbst 1993, 1998), the lesson that public opinion is predominantly manufactured by the media is less appreciated (but see Althaus and Oats-Sargent 2007; Dearing 1989; Groeling and Kernell 1998; Lewis 2001; Lipari 2001). It takes money, personnel, and formal organization to produce public opinion, and the institutions that do so do not conduct polls and disseminate their data out of altruistic concerns for the public interest or for the purpose of furthering democratic values.

Public opinion is more, then, than the channels or dikes for policy makers' actions. Politicians and presidential administrations may well use both publicly available and private survey-research data (and/or focus groups) to help guide their decisions, but we also know that policy makers routinely mischaracterize and manipulate public opinion data for their own ends (Jacobs and Shapiro 2000; Kull and Destler 1999; Sobel 2001). Even worse, as the above example of public opinion on global warming/climate change suggests, public opinion may not even be solicited in the first place and, even should it be solicited, the resultant polling data may be ignored and remain unpublicized. With some exception--such as findings with respect to standard questions--public opinion polling serves as an indistinct and inarticulate voice of the American people.

A second finding about who speaks for the public is the fleeting nature of presidential leadership, notwithstanding the fact that the U.S. president has arguably the strongest claim on representing the public's voice. While wars and foreign policy crises may produce short-lasting rally effects and while presidents may enjoy brief periods of high popularity under conditions of partisan control, landslide electoral victories, and personal appeal, periods of more than a few months of

high public approval of presidential job performance are relatively few: the high approvals of President Bush's job performance from early 2001 up until about mid-2003; President Clinton from late 1997 up until 1999 (at which point his job approval rating stayed at about 60% until he left office); President George H.W. Bush from 1989 until mid-1991; President Reagan from mid-1985 to mid-1986; President Carter through most of 1977; the first year after Nixon took office in 1969; and LBJ's first few months in office at the end of 1963 and first half of 1964 (http:// online.wsj.com/public/resources/documents/info-presapp0605-31.html).

These public-opinion findings, mostly culled from Gallup data, suggest the media's complicity in the creation of the president's status, in fact. The media (and Congress) were very much complicit in the ability of the president to speak for the public. On the one hand, they went along with the Bush White House following the events of September 11, 2001 (Bennett, Lawrence, and Livingston 2007). On the other hand, they undermined President Clinton's ability to speak as the public's voice once the Lewinsky scandal broke, despite the president's high job approval ratings. Similarly, the media undermined the ability of President George H.W. Bush to speak for the public in the year and a half before the liberation of Kuwait of early 1991, notwithstanding Bush's high approval ratings over that same period; instead, the media focused on the president's plan to raise taxes despite his campaign pledge to the contrary (even if the president's actions made sound fiscal sense), on the president's "wimp" factor, and on the GOP losses in the November 1990 midterm elections. Conversely, President Reagan did not enjoy particularly high job approval ratings and had relatively modest average approval ratings (lower than that of the elder Bush and Bill Clinton, though higher than those of Presidents Carter and Nixon). Yet Reagan has been lionized as a master communicator and "Teflon President," while being praised by politicians and pundits on the left and right of the political spectrum. Again we see the complicity of the media.

In short, the president appears to enjoy only brief and ever-shorter political honeymoons upon election into office, where the early, high popularity of presidents Johnson, Nixon, and Carter have not been replicated by subsequent presidents with the exception of the elder Bush. Presidents may enjoy foreign policy rally effects as a result of major military actions, yet even these are hard to sustain if the recent history of the two Bush presidencies are indicative, and less significant foreign policy initiatives such as the U.S. interventions in Somalia, Haiti, and the former Yugoslavia do not necessarily result in high presidential approval ratings. A third finding from this exploration of public leadership, then, is that the media constitute a critical variable with respect to both public opinion per se and presidential leadership.

Yet if the media exert critical influence on when and how public opinion polling data and U.S. presidents are able to speak for the public, it becoming ever less clear exactly what kind of a voice the media have. The media constitute a multitude of voices, even as they are generally more likely to be negative, and indeed cynical, than positive or proactive. Little wonder then at the "vanishing voter" of the 1990s and most of the 2000s, and at the skepticism by which the public views the political process, elected officials, and national government (Patterson 2004; Hibbing and Theiss-Morse 2000). But the welcoming news for a new presidential administration is that, with the breakdown of the established media and many of the traditional reporting norms and practices, thanks to the new media, the mainstream media have lost their lock on setting the public agenda and framing public debate.

The conventional wisdom that used to be propagated by the mainstream media has come under increasing challenged from alternate political voices and differing constructions of social knowledge as a result of the burgeoning number of political Web sites, blogs, and Internet discussion forums. The history of the 2008 electoral campaign is again suggestive of how the mainstream media's agenda setting and news framing in their news coverage and political commentary appear to have become less determinative of public conversations and voters' commitments.

This result may, to be sure, be a function of the severity of the economic and foreign policy problems that now confront the United States and American public. But it is hard--perhaps impossible--to imagine a turn of events in the foreseeable future when the economy and international relations become less serious. And in the circumstances of an increasingly factionalized and unstable media universe, credible policy makers have ever more opportunity to communicate directly with the public and to promote their own visions and answers to the public and to other policy makers.

This is not to say that the established media are irrelevant or unimportant. Far from it. But they are much less dominant than they have been in the past. Just as the printed press of the 1980s and 1990s could not disregard the broadcast media when reporting the news, the print and broadcast media of the 2000s and 2010s cannot disregard the news, commentary, videos, and blogs on the Internet. The result of this may not necessarily be the fuller expression of public opinion per se. Nor will it necessarily be the fuller articulation of the president's voice or those of his party. Yet this new manifold public voice, supplementing that of the established media, may serve as a better representation of the thoughts and feelings of a pluralist and heterogenous American public--thoughts and feelings that we would like to think the political system can reflect.

References

Althaus, Scott L., and Jennifer Oats-Sargent. 2007. Influencing agendas: Untangling the reciprocal effects of polling data and news coverage on the topics of public discourse. Paper presented at the 2007 meeting of the Midwest Political Science Association, April 3.

Almond, Gabriel. 1950. The American people and foreign policy. New York: Harcourt, Brace.

Ceasar, James W., Glen E. Thurow, Jeffrey K. Tulis, and Joseph M. Bessette. 1981. The rise of the rhetorical presidency. Presidential Studies Quarterly 11 : 158-71.

Cohen, Jeffrey E. 2004. If the news is so bad, why are the presidential polls so high? Presidents, the news media, and the mass public in an era of new media. Presidential Studies Quarterly 34: 494-515.

Cook, Timothy E. 1998. Governing with the News: The news media as a political institution. Chicago: University of Chicago Press.

Dearing, James W. 1989. Setting the polling agenda for the issue of AIDS. Public Opinion Quarterly 53: 309-29.

Edelman, Murray. 1977. Political language: Words that succeed and policies that fail. New York: Academic Press.

Edwards, George C. III. 2003. On deaf ears: The limits of the bully pulpit. New Haven, CT: Yale University Press.

Eliasoph, Nina. 1998. Avoiding politics. Cambridge, UK: Cambridge University Press.

Entman, Robert M. 1993. Framing: Toward clarification of a fractured paradigm. Journal of Communication 43(4): 51-58.

--. 2004. Projections of power: Framing news, public opinion, and U.S. foreign policy. Chicago: University of Chicago Press.

Fishkin, James S. 1991. Democracy and deliberation: New directions for democratic reform. New Haven, CT: Yale University Press.

Gamson, William A. 1984. What's news? A game simulation of TV news. New York: Free Press.

Ginsberg, Benjamin. 1986. The captive public: How nass opinion promotes state power. New York: Basic Books.

Groeling, Tim, and Samuel Kernell. 1998. Is network news coverage of the president biased? Journal of Politics 60: 1063-88.

Hart, Roderick P. 1987. The sound of leadership. Chicago: University of Chicago Press.

Herbst, Susan. 1993. Numbered voices: How opinion polling has shaped American politics. Chicago: University of Chicago Press.

--. 1998. Reading public opinion: How political actors view the democratic process. Chicago: University of Chicago Press.

Hibbing, John, and Elizabeth Theiss-Morse. 2000. Stealth democracy. Cambridge, UK: Cambridge University Press.

Iyengar, Shanto. 1991. Is anyone responsible? How television frames political issues. Chicago: University of Chicago Press.

Iyengar, Shanto, and Donald Kinder. 1987. News that matters: Television and American opinion. Chicago: University of Chicago Press.

Iyengar, Shanto, and Jennifer R. McGrady. 2007. Media politics: A citizen's guide. New York: Norton.

Jacobs, Lawrence R., and Robert Y. Shapiro. 2000. Politicians don't pander. Chicago: University of Chicago Press.

Kernell, Samuel. 1997. Going public: New strategies of presidential leadership (3rd ed.). Washington, DC: CQ Press.

Key, V. O. Jr. 1961. Public opinion and American democracy. New York: Knopf.

Kull, Steven, and I.M. Destler. 1999. Misreading the public: The myth of a new isolationism. Washington, DC: Brookings Institution Press.

Lewis, Justin. 2001. Constructing public opinion. New York: Columbia University Press.

Lindsay, James M. 1992-93. Congress and foreign policy: Why the hill matters. Political Science Quarterly 107: 607-28.

Lipari, Lisbeth. 2001. Voice, polling, and the public sphere. In Politics, discourse, and American society: New agendas, edited by R. Hart and B. Sparrow. Lanham, MD: Rowman & Littlefield, 129-49.

McCombs, Maxwell. 2004. Setting the agenda: The mass media and public opinion. Cambridge, MA: Polity.

Medhurst, Martin J., ed. 1996. Beyond the phetorical presidency. College Station: Texas A&M Press.

Mueller, John. 2000. Public opinion as a constraint on U.S. foreign policy: Assessing the perceived value of American and foreign lives. Paper presented at the National Convention of the International Studies Association, Los Angeles, March 14-18.

--. 2003. American public opinion and military ventures abroad: Attention, evaluation, involvement, politics, and the wars of the bushes. Paper presented at 2003 Annual Meeting of the American Political Science Association, Philadelphia, August 28-31.

Neuman, W. Russell, Marion Just, and Anne Crigler. 1992. Common knowledge: News and the construction of meaning. Chicago: University of Chicago Press.

Noelle-Neumann, Elisabeth. 1983. The spiral of silence. Chicago: University of Chicago Press

Ornstein, Norman J., and Thomas E. Mann. 2006. When Congress checks out. Foreign Affairs 85(6).

Page, Benjamin I., and Robert Y. Shapiro. 1993. The rational public. Chicago: University of Chicago Press.

Patterson, Thomas E. 1993. Out of order. New York: Knopf.

--. 2004. The vanishing voter. New York: Knopf.

Peterson, Paul E. 1994. The president's dominance in foreign policy making Political Science Quarterly 109: 215-34.

Raddatz, Martha. 2008. Interview of the vice president by Martha Raddatz, ABC News. http:// www.whitehouse.gov/news/releases/2008/03/20080319-5.html.

Rozell, Mark. 2003. Congress and the media. In Media power media politics, edited by Mark Rozell. Lanham, MD: Rowman & Littlefield.

Schiffer, Adam. 2008. Conditional press influence on politics. Lanham, MD: Lexington Books.

Schraufnagel, Scot, and Stephen M. Shellman. 2001. Research note: The two presidencies, 1984-98: A replication and extension. Presidential Studies Quarterly 31: 699-707.

Sobel, Richard. 2001. The impact of public opinion on U.S. foreign policy since Vietnam. New York: Oxford University Press.

Sparrow, Bartholomew H. 1999. Uncertain guardians: The news media as a political institution. Baltimore: The Johns Hopkins University Press.

--. 2006. Resolved, the media are too hard on presidents: Con. In Debating the presidency: Conflicting perspectives on the American experience, edited by Richard J. Ellis and Michael Nelson. Washington, DC: CQ Press, 68-74.

Sparrow, Bartholomew H., and Natalie Jomini Stroud. 2007. Making public opinion: Polling, the nedia, and the United States' response to 9/11. Paper presented at the annual meeting of the American Political Science Association, Chicago, August 30-September 2.

Tulis, Jeffrey K. 1987. The rhetorical presidency. Princeton, NJ: Princeton University Press.

Wattenberg, Martin P. 2004. The changing presidential media environment. Presidential Studies Quarterly 34: 557-72.

Wildavsky, Aaron. 1966. The two presidencies. Trans-Action 4: 7-14.

(1.) The results using the Factiva database were comparable to those reported above using LexisNexis. Under the Factiva database's category of "Newspapers: All" (which includes foreign newspapers and wires stories): A search for the months of March and April 2008 disclosed 250 news stories on Obama and bowling, 67 stories on Obama and flag pins, 73 stories on Obama and Iran, and 69 stories on Obama and climate change/global warming.

(2.) The search term was "global warming OR climate change W/10 poll OR public opinion."

BARTHOLOMEW H. SPARROW

University of Texas at Austin

Bartholomew H. Sparrow is a professor of government at the University of Texas and the author of The Insular Cases and the Emergence of American Empire and Uncertain Guardians: The News Media as a Political Institution.