Presidential rhetoric and the power of definition.

Zarefsky, David

Presidential rhetoric is studied from the perspectives of both the social sciences and the humanities. From a humanistic perspective, scholars are concerned with the uniqueness of exemplary cases as well as with recurrent patterns, and they seek insight and appreciative understanding more than prediction and control. I study presidential rhetoric in the beliefs that it increasingly is what the presidency is about and that it makes a difference.

These assumptions are directly challenged by the results of George Edwards' extensive research program. In his most recent book, Edwards observes that, although "[engaging] in a prominent campaign for the public's support" has emerged in modern times as the president's "core strategy for governing," still "presidents usually fail in their efforts to move the public to support them and their policies." (1) Edwards is particularly bedeviled by the paradox that, whereas people generally assume that presidential rhetoric makes a difference, "very few studies focus directly on the effect of presidential leadership of opinion...." (2)

Edwards' findings are consistent with the "limited effects" model that has characterized media research for the past 50 years. Whereas people earlier had been concerned that mass media were like a hypodermic needle with which audiences were injected with strong doses of propaganda, the empirical research generally found that mediated messages had little effect at all--so long as one understood "effect" in terms of measurable changes in the audience's beliefs or attitudes. (3) And yet few would argue seriously that mediated messages make no difference, just as few would say that about the specific case of presidential rhetoric. So Edwards' conclusion remains paradoxical.

The research agenda described here attempts, at least in a small way, to move past this paradox and to contribute to answering these questions: (1) How do we understand the nature of presidential rhetoric and its effects? (2) What does presidential rhetoric do?

(3) How do we know?

Understanding Rhetoric and Its Effects

Both terms--"presidential rhetoric" and "effect"--are often understood too narrowly: "presidential rhetoric" as public speeches and "effect" as quantitatively measurable changes in indices of people's attitudes or beliefs. These conceptions reflect an overly simplistic understanding of the process of communication and the nature of rhetorical transactions.

Complicating our understanding of presidential rhetoric is the fact that rhetoric is both a type of evidence available for use by scholars in any discipline, and a field of study in its own right. Social scientists, for example, can draw on presidential speeches as data. They may regard them as independent variables and measure their consequences for opinion and attitude change. Not surprisingly, they may not find any. Why? First, we know from communication research that attitudes are seldom changed on the basis of a single message. Second, replacement of an attitude or opinion with another is only one kind of attitude change. Reinforcement of one's initial position, modification in the salience of a belief or attitude, changes in perception of what other beliefs or attitudes are related to the one at hand, or differences in interpretation of what the belief or attitude means are all examples of other types of change. And third, the focus on the message-audience relationship--looking for the effects of messages on audiences--is only one dimension of a rhetorical transaction, and not always the most helpful or informative. In particular, it tends to reduce the message to a verbal text and then to treat the text as a "black box," rather than seeing its dynamics as interesting and worthy of analysis in their own right.

The field of rhetorical studies, by and large, makes different ontological assumptions and relies on a more complex view of the rhetorical transaction. It emphasizes contingency and choice rather than predictability and control. According to this view, the rhetor (speaker or writer) makes choices, with an audience in mind, about the best way to achieve his or her goals in the context of a specific situation. (4) Those choices--about such matters as argument selection, framing, phrasing, evidence, organization, and style, as well as about staging, choreography, and other aspects of the presidential performance--are embodied in the text that the rhetor composes and the context in which it is delivered. An audience, also influenced by context, perceives this text, interprets it, participates thereby in determining what it means, and is affected by it.

On this reading, there are at least three different dimensions of a rhetorical transaction that may be of scholarly interest. One is the previously mentioned relationship between messages and audiences, and it is largely an empirical matter. Another is the relationship between rhetor and text, which could be approached as a historical matter, exploring archival and other resources in an attempt to discern what the author was thinking and what rhetorical choices he or she knowingly made, and why. Alternatively, one might argue that the rhetor's motives are embedded in the text itself and one might employ modes of inquiry ranging from content analysis to psychoanalysis in an attempt to reveal those motives or choices.

Finally, one might take the text itself as the point of departure, analogous to a work of literary or visual art. (Again, "the text" refers not only to the words the president speaks but to the entirety of the presidential performance. For example, the fact that a State of the Union address is delivered to a joint session of Congress, with the president appearing before a giant U.S. flag in the chamber of the House of Representatives, is as much a part of his speech as are the words he speaks.) In this last case, the key relationship is between the text and the rhetorical critic, who uses different reading strategies to reveal levels of meaning or significance in the text. This is a process of speculative reconstruction of the text, informed by the critic's insight into the text's possibilities.

A brief example might make these three difficult perspectives clearer. Consider President George W. Bush's speech to a joint session of Congress on September 20, 2001. He pronounced war on terrorist organizations of global reach and issued an ultimatum to nations that might harbor terrorists. The first approach, studying message-audience relationships, would ask whether, after the speech, public opinion changed on such matters as support for military action against Afghanistan or the president's overall performance in office. The second approach, examining speaker-message relationships, would study the development of the message, asking such questions as how involved the president was in writing the speech or why the principal writer put God in the text, choosing between freedom and fear--or it might infer the president's world view from his use of the binary that nations either support us or support the terrorists. The third approach, unpacking the text, might observe the open-ended characterization of terrorism and note that the label of "terrorism" could be used similarly by leaders of other nations with reference to their foes, so that the text offers a potentially dangerous invitation for unwanted "copycat" activities by other nations.

Our understanding of presidential rhetoric should not be limited to just one of these perspectives. Rhetoric is not only an alleged cause of shifts in audience attitudes. It is also a reflection of a president's values and world view. And it is also a work of practical art, often richly layered and multivocal, that calls for interpretation.

The picture, though, is messier even than this. The distinctions among "historical," "literary," and "empirical" perspectives are often blurred by rhetorical scholars themselves. They employ causal language and thereby suggest empirical claims when they really do not mean to make causal arguments but have other dimensions of rhetoric in mind. This conceptual sloppiness invites the rejoinder, especially from social scientists, that the rhetorician is making claims unencumbered by evidence, and therefore that no effect can be attributed to presidential rhetoric. A number of rhetorical scholars, myself included, listened uncomfortably as George Edwards offered just such a critique in 1995. (5) His point is well taken. But the remedy is not for rhetorical critics to morph into empirical social scientists, but rather for them to be more precise about what they are claiming and to eschew misleading causal language when it does not fit.

Again, an example may help to illustrate. Suppose a student of President Bush's September 20 address argues that the speech will cause other leaders to repress dissidents while claiming that they are cracking down against terrorists. As a causal proposition, that is highly unlikely, because it attributes incredible agency to the words of the U.S. president and implies that other leaders lack their own decision-making calculus. In any event, it is impossible to prove: How in the world could one control all extraneous factors? The rhetorical critic, we may assume, is not so stupid as to blunder into these traps. It is more likely that the critic really means to say something else. The statement about what the text "causes" actually says something more like this: The September 20 speech justifies a strong response to terrorism, but it leaves terrorism largely undefined. Reasoning by analogy, it is not hard to imagine another leader, besieged by intense foreign opposition or domestic dissent, characterizing that opposition as terrorism and responding the same way as the president of the United States. One perhaps could say that the text "invites" this reading (meaning that a reasonable person plausibly might see it that way) or that a perceptive reader might see this possibility in the text. But this is quite a different matter from saying that the president's rhetoric causes copycat responses by others. This example is typical of presidential rhetoric in that it is far more likely to suggest possibilities and to issue invitations than it is to determine outcomes.

There are yet other complications to explore. Aristotle defined rhetoric as the faculty of discovering the available means of persuasion in the given case. Those last four words emphasize that rhetoric is situational; it is grounded in particulars and resists easy generalization. Unpacking a text, probing its dimensions and possibilities, helps the scholar to understand better the richness of a very specific situation that already has passed and will not return in exactly the same way. But if every rhetorical moment is altogether unique, then our assessments are highly idiosyncratic and have no generalizability. The solution to this conundrum is to acknowledge that, while no two situations are exactly alike, patterns of rhetorical choice do tend to repeat across situations with the same central characteristics. This need not imply a formal set of genres or archetypes, types, (6) but it does suggest that rhetorical masterpieces can be studied in the same way that great works of literature are studied: with an eye both to offering new perspective on the case at hand and to suggesting broader principles that will help to explain rhetorical practice more generally.

Finally, who is the audience for presidential rhetoric? This is not as simple a question as it might seem. Although the essence of the public presidency is the assumption of direct presidential appeals for popular support, we know that even presidents have difficulty gaining a mass audience. (7) Especially in the contemporary era, the primary audience for presidents often is other politicians or the media. The strategies of "going public" may be designed largely to influence the media, for example. (8) The assumption is that media response is important in its own right and that, through such devices as framing, media "translate" presidential messages and influence how they are understood by ordinary citizens.

Presidential Definition

What has been done so far is to create a space for refreshed understanding of rhetoric bur not yet to fill it with conceptual content. Let me therefore advance a claim about what presidential rhetoric does: It defines political reality.

The key assumption I make is that characterizations of social reality are not "given"; they are chosen from among multiple possibilities and hence always could have been otherwise. Whatever characterization prevails will depend on choices made by political actors. People participate actively in shaping and giving meaning to their environment, and they do so primarily by means of naming situations within it. Naming a situation provides the basis for understanding it and determining the appropriate response. (9) Because of his prominent political position and his access to the means of communication, the president, by defining a situation, might be able to shape the context in which events or proposals are viewed by the public. Of course, not all presidential attempts at definition evoke a positive public response, and one test of the effectiveness of presidential definitions is to find evidence of such resonance. (10)

To choose a definition is, in effect, to plead a cause, as if one were advancing a claim and offering support for it. But no explicit claim is offered and no support is provided. The presidential definition is stipulated, offered as if were natural and uncontroversial rather than chosen and contestable. President Bush simply identified the estate tax as the "death tax," for example, or called intact dilation and extraction "partial-birth abortion," or pronounced that rolling back future tax cuts for the wealthy was really a tax increase. One could argue that each of these definitions is right or wrong, but the point is that, in defining the situation, the president makes no explicit argument.

Each of these brief examples illustrates how there are interests at stake in how a situation is framed. The definition of the situation affects what counts as data for or against a proposal, highlights certain elements of the situation for use in arguments and obscures others, influences whether people will notice the situation and how they will handle it, describes causes and identifies remedies, and invites moral judgments about circumstances or individuals. (11) Accordingly, presidential definition resembles what William Riker calls heresthetic: "the art of structuring the world so you can win." (12) Whether the art is practiced consciously or instinctively does not matter. It can be conscious, as when presidential candidates try to "define" their opponents, (13) or instinctive, as when a president gives voice to what may be unexamined ideological commitments.

Inspection of numerous case studies suggests several means by which presidents exercise their power of definition. One is to create associations with other terms, expanding the meaning of a term to cover the new case at hand. September 11 was defined as "war" by linking it to the specific attributes of that term that were indisputably present in the situation, thereby extending the reach of the term. The argument is one of analogy, yet no explicit analogy was voiced.

Second, a situation can be defined by dissociation. (14) This consists of breaking a concept into parts in order to identify one's proposal with the more favored part. One prefers the spirit of the law over the letter, the real over the apparent, practice over theory, and so on. When President Kennedy identified his arms control programs with "real peace," not just the temporary absence of military conflict, he was engaged in a dissociation. The concept of "peace" was taken apart and reconstructed.

Third, a situation can be defined by identifying it with one or more condensation symbols. (15) These are symbols which designate no clear referent but "condense" a host of different meanings and connotations that otherwise might diverge. They are particularly useful in defining an ambiguous situation because people can highlight different aspects of the symbol yet reach the same conclusion. For example, President Clinton's approach to the budget surpluses of the late 1990s was "Save Social Security first." Saving Social Security is a theme with positive resonance, even though people mean different things by it. The symbol of saving this cherished program gathers support from among people who may have different reasons for offering it and who may mean different things by it. The power of the definition is its ability to condense divergent emotional reactions.

Fourth, presidents can rely on frame shifting, postulating a different frame of reference from the one in which the subject normally is viewed. The effect is that people see the thing "in a different light" and their attitudes about it therefore change. Riker, in fact, suggests that "most of the great shifts of political life result from introducing a new dimension." (16) President Bush employed frame shifting in his ex post facto justification for the 2003 war in Iraq. When no weapons of mass destruction were found, he invited listeners to see the war from the perspective of the benefits of eliminating a tyrant, even though that had not been the original justification, rather than from the frame of protecting the United States and other nations against the risk of biological, chemical, or nuclear weapons.

These four approaches undoubtedly do not exhaust the types and means of rhetorical definition. Part of the reason for amassing data on specific cases is that it is a way to broaden awareness and deepen understanding of the resources available to presidents as they engage in public persuasion.

Eight Case Studies

Several cases of presidential use of the power to define will help to show how this rhetorical power can be used to alter public conceptions of political reality, thereby shifting the ground--though not always in the president's favor.

Case One, When enforcement of an excise tax on whisky triggered taxpayers' protest in western Pennsylvania in 1794, George Washington interpreted these events within the frame of Shays's rebellion in Massachusetts eight years before. To the president, the protest was not a case of intimidation and violence at the local level, to be met by the law enforcement authorities, but a fundamental challenge to the authority of the national government under the Constitution. Defining the situation in this way, Washington deemed it important to convey the message that the new government would not be cowed by threats to it or to him. Accordingly, he issued a request for militia volunteers to go to western Pennsylvania to quell the disturbance. When they found no menace in the West, the president interpreted this as evidence of the deterrent power of federal force. These definitional moves helped him to make his point, but he then overreacted in his 1794 Annual Message, claiming the protests to have been caused by the pro-French democratic societies. The resulting criticism served to topple Washington from the pedestal of nonpartisanship, so that he was seen as a Hamiltonian Federalist and was publicly vilified by Republicans during the last years of his term. (17)

Case Two. Until 1832, presidential elections were understood as a process for selecting the man who would be most capable to administer the Executive Branch. They chose the office holder but not his policies. Near the end of his first term, however, Andrew Jackson was the target of National Republicans in Congress who sought to embarrass him by voting to recharter the Bank of the United States (which he opposed) four years before its charter was to expire. Jackson vetoed the Bank bill and, in an emotional message to Congress, called on voters to sustain his action in the coming presidential election. He was indeed reelected, although it is likely that this result was independent of the Bank issue. (18) Nevertheless, Jackson defined his reelection as a mandate to get rid of the Bank. He said, "Whatever may be the opinions of others, the president considers his reelection as a decision of the people against the bank.... He was sustained by a just people, and he desires to evince his gratitude by carrying into effect their decision so far as it depends upon him," (19) and he engineered the Bank's collapse by ordering withdrawal of government deposits. From then on, winning candidates often interpreted their election--no matter how ambiguous the circumstances--as a mandate for particular actions. John Tyler did so in 1844, defining Polk's election as a mandate for the immediate annexation of Texas. Abraham Lincoln did so in 1860, citing the election results to justify his refusal to engage in last-ditch compromise moves to avert the secession crisis. Grover Cleveland did so in 1892, viewing the election results as a mandate to abandon silver in favor of the gold standard. Warren G. Harding did so in 1920, defining the results as a mandate to stay out of the League of Nations. Lyndon Johnson did so in 1964, understanding the results as a mandate for the enactment of Medicare. Ronald Reagan did so in 1980, finding in the results a mandate for substantial tax cuts. George W. Bush did so in 2000, viewing his election, albeit disputed, as a mandate to enact the Republicans' conservative platform rather than moderate measures in the name of national unity. (20)

Case Three. Abraham Lincoln accepted civil war in order to preserve the Union by demonstrating the impossibility of secession. Although personally opposed to slavery, he insisted that he had no power to alter it where it already existed. For this reason he was willing even to see a constitutional amendment that would have protected slavery in the southern states in perpetuity. The war was about the integrity of the Union, not about slavery. But Lincoln, like most of his countrymen, had not expected a long and protracted war. As Union losses continued to mount with no end to the struggle in sight, it became harder to rally support and to inspire sacrifice for the sake of the status quo ante. Moreover, restoration of the Union was a morally neutral principle, unlikely to deter Britain and France from recognizing the Confederacy as a legitimate government, perhaps even intervening on its behalf.

In a series of rhetorical moves during 1862, Lincoln gradually redefined the aims of the war. He did not abandon his commitment to colonization of freed blacks as the optimal policy, but in his 1862 Annual Message he began to suggest that the opposition to emancipation was based on arguments that were "imaginary, if not sometimes malicious." (21) He issued the Emancipation Proclamation, which was carefully drawn to apply only to areas that were in rebellion--or, in other words, to the areas where he had no power to enforce it. By doing so, he was able to define emancipation not as a new objective but as a means to preserve the Union--by inducing slaves in rebellious states to desert their masters and enlist in the Union army, by dissuading Britain and France from intervention, and by arousing Northerners to sacrifice for the sake of a moral good. He was able to bypass challenges to his authority through defining a radical change in policy as continuity, as merely a means to achieve the already established goal.

Case Four. Traditionally, liberalism had been understood as implying minimal government intervention in the life of the individual. Government had no role to play in economic development or social welfare; individuals would fend for themselves in the race of life. Franklin D. Roosevelt effectively redefined liberalism as meaning just the opposite. In speeches during the 1932 campaign and in his actions thereafter, he did so by arguing that individual freedom was threatened by the consolidated power of big business. Individuals could not bargain freely in the marketplace when power relations were so unbalanced. Accordingly, protecting individual freedom required establishing a countervailing power to offset the weight of big business, and this function would be performed by big government. It is hard to remember, but before Presidents Reagan and Bush redefined liberalism once again, the view of an activist government was often positive, because it had the energy to improve the human condition. Even Republican Presidents Eisenhower and Nixon felt constrained not to dismantle the major programs of the New Deal and its progeny. In this way, major government programs achieved a national consensus even though they were understood as programs of liberal reform. Defining the new as a means to achieve the old, Roosevelt transformed the political landscape for nearly half a century. (22)

Case Five. Lyndon Johnson entered the presidency needing quickly to demonstrate his liberal credentials to a public that did not yet know him very well. Meeting with his economic advisers, he learned that John E Kennedy had been thinking about an antipoverty initiative as part of his legislative program. Johnson, believing that an antipoverty effort was the natural successor to the New Deal, signed on at once. He announced his commitment in the 1964 State of the Union message, proclaiming that "this administration, here and now, declares unconditional war on poverty." The war metaphor was chosen deliberately, Johnson indicated, in the belief that it would rally the people and mobilize support. (23) These were important goals, particularly because the antipoverty legislation was proposed in response to no public groundswell and in the face of skepticism about whether poverty ever could be cured. The metaphor was not just a rhetorical flourish; it framed the new initiative in a favorable way. It partook of the habits of thought that characterized the crisis presidency under the impact of the Cold War. (24) Simply put, a crisis (such as war) rearranges the rhetorical ground. The urgency of the situation requires quick response and establishes a presumption in favor of action. There is no time to consider carefully all the arguments and objections that might arise during peacetime. So debate is truncated, Congress gives its blessing without much understanding of the details, and the president takes on the persona of the commander-in-chief. These moves all gave Johnson advantages he would not have had under conditions of normal politics. They hastened the adoption of the War on Poverty--although they also would create problems for its subsequent implementation. (25)

Case Six. This case also involves Lyndon Johnson. In a speech at Howard University in 1965, he effectively redefined "equal opportunity" to embrace equal outcomes, not just equal chances. He did so by analogy to a foot race, saying, "You do not take a person who, for years, has been hobbled by chains, liberate him, bring him to the starting line of a race, saying 'you are free to compete with all the others,' and still justly believe you have been completely fair." (26) He distinguished between apparent and real equality of opportunity, the former represented by bringing the contestants to the same starting line and the latter by compensating for disadvantages suffered by any of the contestants. The goal, the president proclaimed, was "not just equality as a right and a theory but equality as a fact and a result." Johnson was engaging in dissociation--taking a concept that presumably has a single, clear meaning, dividing it into separate aspects one more favorable than the other, and equating the antagonist's position with the dis-preferred term and one's own with the preferred term.

I do not mean to suggest that Lyndon Johnson would have endorsed every idea subsequently developed under the rubric of affirmative action. He most likely was thinking of compensatory education and training, not numerical goals or quotas. The point, however, is that his redefinition of "equal opportunity" created the rhetorical space that made affirmative action possible, by identifying a new policy concept with an established and accepted value. (27)

Case Seven. Ronald Reagan also employed dissociation. He sought to reduce welfare programs without seeming heartless. He described isolated but egregious cases of welfare fraud, distinguished the perpetrators from the "truly needy," and claimed that his reforms would maintain a "safety net" for the truly needy without squandering public resources on ne'er-do-wells who did not deserve it. Reagan took the previously unitary category of "needy" and divided it in two, maintaining his popularity even as he challenged welfare programs. He had effectively redefined the nature of welfare. (28)

Case Eight. The final example returns to the metaphor of war, as it was deployed in response to the terrorist attacks of September 11,2001. President Bush instantly and instinctively reacted to news of the attack by saying simply, "We are at war." He employed the same definition of the situation in his September 20 speech and he has maintained it ever since. Only in his 2004 State of the Union address did he find it necessary even to respond to unnamed critics who, he said, contested this definition of the situation. But calling the terrorist attacks "war" is hardly self-evident. They had some of the attributes normally associated with war: Attacks were launched and lives were lost. But they lacked other characteristics: The attack was not military; it did not come from a nation state; no country declared war on us, nor did we on any other nation. Yet defining the situation as war helped to clarify what responses were appropriate: national unity, quick response without debate or deliberation, rallying around the president, overt displays of patriotism and national pride. All of these responses were evoked by President Bush's characterization of the nature of the terrorist attacks. (29)

These eight examples come mostly from cases examined in my own research. They illustrate what is involved in the use of the presidential power of definition through the performance of presidential rhetoric. This activity often makes a difference. To be sure, however, not all cases of presidential definition are successful in shaping public understanding of a situation. Though he tried, President Carter was unable to enlist the imagery of war on behalf of his energy program, just as President Reagan was unable to change public understanding of the MX missile by his choice to refer to it as the "Peacekeeper." In the current environment, President Bush is having mixed results with his attempt to define failure to extend temporary tax cuts as tax increases. We need to know more about whether there are factors consistently associated with successful or unsuccessful exercise of the power of presidential definition.

Public opinion polls and other empirical surveys of audience response are not likely to measure the effectiveness of presidential definitions, because the definition affects the whole frame of reference within which the question is discussed. Caught up in the shifting frame, people are likely to see it as natural reality rather than the product of rhetorical choice. (Few on September 11 questioned whether we were really at war; only later did it become clear that this was a rhetorical reconstruction.) Truly significant outcomes of presidential rhetoric may pass unnoticed until long after the fact.

How Do We Know?

How then do we know what presidential rhetoric does? At least a few possibilities suggest themselves. One is to see how widely the presidential definition is picked up and used by others. We know, for example, that the military metaphor in the War on Poverty gave rise to numerous collateral metaphors: enlisting for the duration, field generals, weapons and ammunition, victories and defeats. (30) We know that the metaphor was adopted widely and used by others in their discourse. Similarly, we know that the application of the war metaphor to the terrorist attacks found immediate and widespread acceptance. In cases such as these, it seems plausible to suggest that presidential definition has altered the rhetorical landscape by changing the terms in which people think about an issue. In short, we might examine the diffusion of rhetorical constructions, what Ernest Bormann in a different context has called the phenomenon of "chaining out" of a metaphor or image. (31)

More generally, perhaps the appropriate test is that of the historically sensitive researcher who gathers evidence, conducts thought experiments, and advances arguments. Evidence of presidential definition can be found in the texts of public statements, the audio and video records of presidential performance, comments by the president or his aides about his purposes, and the informed speculation of commentators. Evidence about shifts in public understanding include the repetition and chaining out of the definition as discussed above. Thought experiments involve careful questions about counterfactuals. Does it seem reasonable, for instance, that general public understanding of liberalism is different since the New Deal from what it was before? If so, is the shift consistent with Roosevelt's own redefinition? Can one imagine the shift in understanding taking place in the absence of presidential redefinition? Is there any more persuasive explanation for the shift than the hypothesized presidential definition? These questions cannot be answered conclusively, or even at the 05 level of confidence; they are productive of arguments, which are assessed through the exercise of judgment. Because judgment is the object of rhetoric, rhetorical scholars should be comfortable with this approach. It might enrich the understanding of political scientists as well.

Taken together, then, the issues raised here suggest a full research agenda. It includes refining our understanding of how rhetorical definition is performed, examining many more cases of presidential redefinition, searching for factors associated with success or failure, and specifying how we know that frames of reference have been modified. This agenda of research tasks should contribute to the larger goal of moving beyond the impasse in the extant literature and being able to answer the question, "What does presidential rhetoric do?"

(1.) George C. Edwards III, On Deaf Ears: The Limits of the Bully Pulpit (New Haven: Yale University Press, 2003), ix.

(2.) Ibid., 26.

(3.) For a recent statement of this position, see Michael Schudson, The Sociology of News (New York: W. W. Norton, 2003), 19.

(4.) For a classic statement on the situational character of rhetoric, see Lloyd F. Bitzer, "The Rhetorical Situation," Philosophy and Rhetoric 1 (Winter 1968): 1-14.

(5.) George C. Edwards III, "Presidential Rhetoric: What Difference Does It Make?" in Beyond the Rhetorical Presidecy, edited by Martin J. Medhurst (College Station: Texas A&M University Press, 1996).

(6.) Some scholars, however, have tried to identify genres of presidential rhetoric. See, for example, Karlyn Kohrs Campbell and Kathleen Hall Jamieson, Deeds Done in Words: Presidential Rhetoric and the Genres of Governance (Chicago: University of Chicago Press, 1990). Other scholars explore patterns at the microlevel of presidential discourse. For example, see Roderick P. Hart, The Sound of Leadership: Presidential Communication in the Modern Age (Chicago: University of Chicago Press, 1987).

(7.) On this point, see Edwards, On Deaf Ears, 187-217.

(8.) For a strong example of a presidential message for which the media were the target audience, see Kathleen J. Turner, Lyndon Johnson's Dual War: Vietnam and the Press (Chicago: University of Chicago Press, 1985).

(9.) These aspects of definition are explored more fully through a series of case studies in Edward Schiappa, Defining Reality (Carbondale: Southern Illinois University Press, 2003).

(10.) For a fuller explication of these ideas, see David Zarefsky, President Johnson's War on Poverty: Rhetoric and History (University: University of Alabama Press, 1986), 1-11.

(11.) See David Zarefsky, "Definitions," in Argument in a Time of Change: Proceedings of the 10th NCA/AFA Summer Conference on Argumentation (Annandale, VA: National Communication Association, 1998), 4-9; J. Robert Cox, "Argument and the 'Definition of the Situation,'" Central States Speech Journal 32 (Fall 1981): 197-205; Robert M. Entman, "Framing: Toward Clarification of a Fractured Paradigm," Journal of Communication 43 (Autumn 1993): 52-56; Schiappa, Defining Reality; and Schudson, The Sociology of News, 35-47.

(12.) William H. Riker, The Art of Political Manipulation (New Haven: Yale University Press, 1986), ix.

(13.) For example, see Mark Memmott, "Candidates Pick Words that Work in Definition Game," USA Today, March 19, 2004, p. 6A.

(14.) See Chaim Perelman and L. Olbrechts-Tyteca, The New Rhetoric: A Treatise on Argumentation, translated by John Wilkinson and Purcell Weaver (1958; reprinted by Notre Dame: University of Notre Dame Press, 1974), 411-59.

(15.) Edward Sapir, "Symbolism," in Encyclopaedia of the Social Sciences, edited by Edwin R. A. Seligman (New York: Macmillan, 1934), 492, explains condensation symbols.

(16.) Riker, The Art of Political Manipulation, 151.

(17.) These events are described in Thomas P. Slaughter, The Whiskey Rebellion: Frontier Epilogue to the American Revolution (New York: Oxford University Press, 1986); and John C. Miller, The Federalist Era, 1789-1801 (New York: Harper and Row, 1960), 155-62.

(18.) This is the judgment of Jackson's biographer, Robert V. Remini. See Andrew Jackson and the Bank War (New York: W. W. Norton, 1967), 106.

(19.) James D. Richardson, ed. A Compilation of Messages and Papers of the Presidents, 1789-1897, 10 vols. (Washington, DC: Government Printing Office, 1897), 3: 7.

(20.) On election results as mandates and the specific case of Jackson and the Bank, see David Zarefsky, "The Presidency Has Always Been a Place for Rhetorical Leadership," in The Presidency and Rhetorical Leadership, edited by Leroy G. Dorsey (College Station: Texas A&M University Press, 2002), 28-30; and Richard J. Ellis and Stephen Kirk, "Jefferson, Jackson, and the Origins of the Presidential Mandate," in Speaking to the People: The Rhetorical Presidency in Historical Perspective, edited by Richard J. Ellis (Amherst: University of Massachusetts Press, 1998).

(21.) The text of the 1862 Annual Message is in The Collected Works of Abraham Lincoln, edited by Roy P. Basler, 9 vols. (New Brunswick, NJ: Rutgers University Press, 1953), 5: 518-37. For an analysis, see David Zarefsky, "Lincoln's 1862 Annual Message: A Paradigm of Rhetorical Leadership," Rhetoric & Public Affairs 3 (Spring 2000): 5-14.

(22.) On Roosevelt's redefinition of "liberal," see Sidney M. Milkis, "Franklin D. Roosevelt, Progressivism, and the Limits of Popular Leadership," in Speaking to the People, edited by Ellis, 189; Alan Brinkley, Liberalism and Its Discontents (Cambridge: Harvard University Press, 1998), 1-62, passim; and Eric Foner, The Story of American Freedom (New York: Knopf, 1998), 204.

(23.) See Lyndon Baines Johnson, The Vantage Point: Perspectives of the Presidency, 1963-1969 (New York: Holt, Rinehart, and Winston, 1971), 74.

(24.) On the emergence of the crisis presidency, I am indebted to Benjamin Ponder, "The State of Emergency: Global War and the Rise of the Crisis Presidency," unpublished paper, Northwestern University, Fall 2002.

(25.) I develop these themes in Zarefsky, President Johnson's War on Poverty, passim.

(26.) The text of the speech is found in Public Papers of the Presidents: Lyndon B. Johnson, 1965 (Washington, DC: U.S. Government Printing Office, 1966), 2: 635-40.

(27.) For a fuller exploration of these themes, see David Zarefsky, "Lyndon Johnson Redefines 'Equal Opportunity': The Beginnings of Affirmative Action," Central States Speech Journal 31 (Summer 1980), 85-94.

(28.) On this topic, see David Zarefsky, Carol Miller-Tutzauer, and Frank E. Tutzauer, "Reagan's Safety Net for the Truly Needy: The Rhetorical Uses of Definition," Central States Speech Journal 35 (Summer 1984), 113-19.

(29.) For an analysis of this speech, see David Zarefsky, "George W. Bush Discovers Rhetoric: September 20, 2001 and the U.S. Response to Terrorism," in The Ethos of Rhetoric, edited by Michael J. Hyde (Columbia: University of South Carolina Press, 2004), 136-55.

(30. Zarefsky, President Johnson's War on Poverty, 28-29.

(31.) Ernest G. Bormann, "Fantasy and Rhetorical Vision: The Rhetorical Criticism of Social Reality," Quarterly Journal of Speech 58 (December 1972): 396-407. But see G. P. Mohrmann, "An Essay on Fantasy Theme Criticism," Quarterly Journal of Speech, 68 (May 1982): 109-32, for a critical view of the assumptions and procedures underlying this approach.

David Zarefsky is Owen L. Coon Professor of Communication Studies at Northwestern University. He is the author of President Johnson's War on Poverty: Rhetoric and History; Lincoln, Douglas, and Slavery: In the Crucible of Public Debate; and numerous articles on presidential rhetoric.