7

A Guard Dog Perspective on the Role of Media

by George A. Donohue, Phillip J. Tichenor, and Clarice N. Olien, University of Minnesota

A guard dog perspective is offered as a way to better understand the functioning of the mass media as an important set of communication agencies in the social system. mis perspective takes into account the varying roleperformances by mass media in reporting on majorpublic issues. m e guard dog metaphor suggests that media perform as a sentry not for the community as a whole, but for groups having sufficient power and injluence to create and control their own security systems. mis conception is delineated from otherperspectives, which include (aj the traditional fourth estate role of watchdog media, (b) the lapdog view of submissive media, and (c) the view of media aspart of a power oligarchy. Several hypotheses that may be derived for testing the utility of the guard dog perspective are suggested and discussed in light of various bodies of evidence available.

The role of mass media as an independent and powerful force in society is an enduring and controversial issue (Dennis & Merrill, 1991). The literature con- tains a wide range of views about the existence and degree of media power and its autonomy from other powers in the system.

The purpose of this paper is to offer a guard dog perspective of the social role of mass media, and to delineate this role from other perspectives, which include: (a) the traditional fourth estate role of independently powerful watch- dog media; (b) the lapdog view of media as largely submissive to status quo political and economic authority, and (c) a view of media as neither watchdogs nor lapdogs but as part of a power oligarchy in the system.

The guard dog conception takes into account the varying role performances by mass media in reporting on major public issues. Furthermore, it is a perspec- tive from which several hypotheses can be derived for testing its utility in explaining media behavior.

George A. Donohue is a professor emeritus of Sociology, Phillip J. Tichenor is a professor emeritus of Journalism and Mass Communication, and Clarice N. Olien is a professor emeritus of Extension and Rural Sociology, all at the University of Minnesota.

Copyright 0 1995 Journal of Communication 45(2), Spring. 0021-9916/95/$5.00

115

Research Articles

Journal of Communication, Spring 1995

The guard dog metaphor suggests that media perform as a sentry not for the community as a whole, but for those particular groups who have the power and influence to create and control their own security systems. The guard dog media are conditioned to be suspicious of all potential intruders, and they occa- sionally sound the alarm for reasons that individuals in the master household, that is, the authority structure, can neither understand nor prevent. These occa- sions occur primarily when authority within the structure is divided.

Conflict is reported, but in a constrained way and only on certain issues and under certain structural conditions. Given their orientation toward power, media would be expected to concentrate their attention, positive or negative, on the dominant groups. Thus, Dreier (1983) finds the media giving primary attention to the conservative middle of the U.S. spectrum while reporting on the differences within that middle.

Actions of a guard dog are shaped according to (a) the nature of the struc- ture being served and (b) who is defined as the intruder or threat. In the con- sensus atmosphere of a highly homogeneous community, the local power establishment is protected at all times, and media, as part of the structure, pro- tect and maintain this establishment. However much tension there may be at an individual level, open and public conflicts are minimized in all public bod- ies, whether it be a city council, a county board, a school board, or an art coun- cil. Along with other agencies and institutions, the media in such a community rarely report conflict, and the investigative reports that occasionally appear in metropolitan media are entirely absent. If the town is entirely dependent upon agriculture or a single manufacturing firm, the newspaper operates much as the newsletter of a large medical clinic, that is, as a house organ. With few if any visible threats and no challenge, the guardian media in such a structure raise alarms only in unusual circumstances. In the consensus atmosphere, the media are sleeping guard dogs.

Guard dog reporting in such a consensus community appears when external forces present a threat to the local leadership. The newspaper in a small town may editorialize against federal legislation that goes against the grain of power- ful local groups, especially if that legislation promises to alter the local power balance. Rural newspaper opposition to Medicare legislation in the 1960s is a case in point, particularly in rural areas with high populations of aged poor who stood to benefit from Medicare, but were not influential groups in their communities. The community newspapers, at least in Kentucky, were more likely than those in metropolitan areas to accept the professional physician’s opposition at that time to such legislation (Donahew, 1967; Tichenor, Dono- hue, & Olien, 1980).

According to this interpretation, these newspapers were oriented more to concerns of the local medical leadership than to those of the aged population that, if surveyed, may very well have supported the bill. Where different local groups have conflicting interests, the media are more likely to reflect the views of the more powerful groups. Donahew found news coverage to be more nega- tive toward Medicare in small towns (where higher proportions of a lower income population were on old age assistance), and to be unrelated to the pub-

116

Journal of Communication, Spring 1995

lisher’s estimate of community opinion on this issue. In discussing these find- ings, Donahew suggested that it would be worthwhile to study publishers’ meetings with local pressure groups in order to account for more variance in display and direction of coverage.

Through this warning of outside threats, the community press serves as a guard dog for dominant local powers and interests in ways that illustrate the functions of externally based conflict for reinforcement of local cohesion (Coser, 1956). A 19th century rural newspaper could editorialize robustly against the railroads headquartered in New York, Chicago, or Minneapolis and find widespread approval locally for doing so. More recent community press reaction to large discount retail centers in rural areas has often had a similar impact. Leaders’ views that “if we lose Main Street, we lose the town” have been reported frequently, even though a new retail center itself may be readily accepted by the populace as a consumer benefit.

A discount store outside the town boundaries is primarily a challenge to the traditional main street business group that has historically controlled the busi- ness and local governmental decisions in small towns. For the consuming pub- lic, having or not having a traditional main street of entrepreneur owners may or may not be a matter of concern. Few public opinion surveys deal with this question, but the community press generally views saving Main Street as some- thing that the whole community supports.

This conflict-reporting aspect of the guard dog phenomenon occurs in the larger society as well. In reporting foreign affairs, the tendency is to present cri- ses in a framework that is consistent with U.S. foreign policy toward the coun- try in question (Chang, 1989). Herman and Chomsky (1988) find U.S. media coverage biased toward support of the El Salvador government and against the Sandinista government of Nicaragua, much in line with U.S. official policy at the time. Russia bashing, once popular in U.S. media, was replaced t o some extent from 1991-1992 by Saddam Hussein bashing, as foreign policy concerns of the United States underwent change.

It may seem that in supporting official policy, the media are primarily follow- ing popular opinion. Although popular reaction may well have an impact on media in certain cases, the creation of influential public opinion is a process of organization of publics or interest groups that deal with relatively few issues that are decided by legislative bodies and governmental agencies (Blumer, 1954; Key, 1961; Lippmann, 1922). These organized interest groups may have impact on the media, but only to the extent that they receive legitimacy in the political realm.

A case in point is civil rights legislation, which has been attributed to the mobilization effort as well as to public opinion (Burstein, 1979). Some analyses show substantial correlation between constituent opinion and congressional votes (Herrera, Herrera, & Smith, 1992), although the strength of the link varies across issues and it is often pointed out that the relationship may be reciprocal (Page, Shapiro, Gronke, & Rosenberg, 1984). On issues such as school prayer and gun control, federal legislation for long periods has not been supportive of views from public opinion polls (Erskine, 1972; Green & Guth, 1989). Recent

117

Journal of Communication, Spring 1995

passage of the Brady bill illustrates that popular support for an idea typically has political impact only if there are well-organized lobbying efforts behind a proposal. Once those organized forces become established, they receive media attention.’

What Kind of Canine?

The guard dog conception differs substantially from the historically held notion of a fourth estate, which is sometimes attributed to Edmund Burke as he con- sidered the 18th century English Parliament:

n e r e are three estates in Parliament, but in the reporters’gallery yonder sits a fourth estate more important than they all. It is not afisure of speech or witty saying; it is a literal fact, ve y momentous to us in these times. (Carlyle, 1841)

Such an idea of press surveillance has been basic to the ideology of popular and representative government, because it springs from the idea of the popu- lace as sovereign and entering into a social contract with a governing establish- ment that will serve popular interests. It includes the idea of press as “freedom- seeker-and-defender’’ and “forum for the people” (Merrill & Lowenstein, 1971, pp. 98-99) and as investigator as well as adversary of government (Hage, Den- nis, Ismach, & Hartgen, 1976; Schulte, 1981).

Such traditional newspaper names as Sentinel, Monitor, Inquirer, Ehaminer, Observer, Intelligencer, and Advocate seem to spring directly from the fourth estate press ideology. Even Tribune traces back to a Roman-era term that origi- nally meant “champion of the people.” This conception of an independently powerful press with citizens’ interests at heart has been popularized in scores of movies about journalists (Rossell, 1978; Vaughn & Evensen, 1991) and is often reinforced by certain incidents, such as publication of the Pentagon Papers over White House objections (Small, 1972).

The essential ingredients of the fourth estate watchdog perspective include substantial autonomy for the media, their representation of the interests of the populace rather than the dominant groups, and their independent power to directly and independently challenge those dominant groups.

The guard dog approach differs from this perspective on all counts. It rejects the idea of media being autonomous or apart from the system. It assumes that media reflect interests of the dominant groups and the system and have neither

We hypothesize that media coverage of gun control would have become more favorable toward this legislation after the organization of the Brady Bill groups and their various campaigns and lob- bying efforts. When popular opinion and political groups largely agree, media coverage supports them both. When their views diverge, it is typically because the most powerful lobby groups repre- sent something other than popular interests and there is an absence of mobilized groups to advance a countering view. In such cases, coverage will tend to favor the organized political groups.

118

Journal of Communication, Spring 1995

the inclination nor the power to challenge those dominant groups, unless they are already under challenge by other forces.

This perspective also suggests that the fourth estate watchdog role, as tradi- tionally defined, is not only unachieved but is fundamentally unrealistic. There is rarely a single public interest, but a configuration of organized interests that vary greatly in power. Media emphasis is upon maintenance of power relation- ships, that is, power structures. Power structure refers to a configuration of roles and groups that has the capacity to make decisions on a community, regional or societal level.

cal action, but must necessarily deal with ideas and actions generated among the powerful groups. If these groups and agencies are concerned primarily with an external threat, that will be the agenda of the media, as during the McCarthyite period of the 1950s and the ensuing Cold War. When the concern is with internal power struggles, the media will concentrate on those struggles, as during Watergate, the 1992 elections, tax equity, the abortion debate, and women’s rights and health care.

ers, from the guard dog perspective is role playing determined by the circum- stances. Questions asked of authority figures on Nightline or MacNeil-Lehrer are primarily those posed previously by contending powers, such as hostile members of Congress or well-established interest and lobby groups. A typical lead-in by a TV interviewer facing a presidential cabinet member is “How do you respond to the criticism from the (opposing party) that your policies will backfire?”

cial interests, while holding to the potential for a watchdog role. Lippmann (1914/1961, 19221, for example, concentrated on realizing the potential of the press for democratic government. The Hutchins Commission report (Hohen- berg, 1973; Hutchins, 1947) saw the need for the press to emphasize social responsibility, a theme pursued by Schramm (1957) and Altschull (1984).

often implied and sometimes expressed directly in the literature on media agenda-setting (McCombs, 1992; McCombs & Shaw, 19931, which appears to assume independent media power while not necessarily holding to the tradi- tional fourth estate ideal of media acting as a sentry for the larger public.

As an illustration of this, McCombs (1992, p. 817) interprets an earlier study as concluding that ‘ ‘me New York Times put cocaine on the national media agenda,” and makes the more general statement that “attitudes and behaviors” of journalists “are the ultimate filters shaping the nature of the news agenda.” Similarly, in one of the more frequently cited recent studies on agenda-setting, news values of journalists have been said to “shape the prism through which we view much of the wor ld (Rogers, Dearing, & Bregman, 1993; Weaver, Graber, McCombs, & Eyall, 1981, p. 31). After analyzing the 1976 elections, Weaver et al. contend that media possess a kingmaker capability to make or

Media do not have the capacity to innovate in devising social policy or politi-

What seems to be sharp questioning of high governmental officials by report-

Some analysts have, in effect, found media power to be guard dogs for spe-

The guard dog conception should also be distinguished from the media role

119

Journal of Communication, Spring 1995

break presidential hopefuls, a capability so powerful that candidates adjust their campaigns accordingly (p. 207).

This extreme version of the agenda-setting perspective directly responds to the question of who initiates issue agendas in society by answering: the media. This contrasts sharply with the guard dog perspective and the underlying con- flict model of how agendas are set. Institutionalized conflict relationships are pervasive in modern, bureaucratized societies (Dahrendorf, 1959). It is these relationships that establish the definitions of issues, and therefore issue agen- das, which become the focus of organized attention and are reflected in press coverage of these agendas.

Media as Lapdogs

Another view that contrasts sharply with the fourth estate perspective is the idea that media are, or may possibly be, little more than lapdogs (Times-Mirror Corporation, 1986). A lapdog perspective is a total rejection of the Fourth Estate view on all counts. Basic elements of the lapdog view are total submis- siveness to authority, total lack of independent power, obliviousness to all inter- ests except those of powerful groups, and framing all issues according to the perspectives of the highest powers in the system.

Mills (1956) regards the press as “at the disposal of the elites of wealth and power” (p. 315). Similar views, especially on the power of advertisers over the media, have been expressed by Altschull (1984), Bagdikian (1983), Breed (1958) and a number of British writers (Curran, 1977; Golding, 1986; Gurevitch & Blumler; Morley, 1976).

Although the guard dog conception is based on deference to authority, it dif- fers from the lapdog version in that it does assume a conflict role of media, one that would not necessarily produce abject subservience. In the lapdog view, the only media role in conflict would be the defense of the powerful against outside intruders, whereas in the guard dog perspective occasions would arise that entail reporting of conflicts between dominant powers or power blocs.

Deference and outright submissiveness are not the same. Although there is considerable evidence that reporters depend heavily upon powerful political and administrative leadership (Chang, 1989; Molotch & Lester, 1975; Shepherd, 1981; Sigal, 1973), this does not mean that the reporters are always passive, although passivity may occur in some situations. Degree of deference depends upon the nature of community structure as well as on the concentration of power in the larger society. Edelstein and Schulz (1963) found leaders in a small community favoring deference by the press to community leadership before discussing controversies, rather than editorial initiative in reporting such discussion.

In an extensive case study, Paletz, Reichert, and McIntyre (1971) identified a number of mechanisms through which local print media may reinforce the

120

Symposium / ‘!Guard Dog” Journalism

authority of local governing bodies. Community editors in Minnesota mini- mized reporting of conflict in the interest of protection of community institu- tions, particularly in the consensus system of more homogeneous communities. This minimization of conflict was especially apparent in weekly newspapers whose editors were part of the local power structure (Olien, Donohue, & Tichenor, 1968).

ing is far more frequent. More pluralistic structures are more formalized and interactions generally operate according to a conflict model. Public issues arise from the interactions that occur among powerful agencies as a matter of nor- mal, and generally formalized, practice. In these structures, all roles tend to be specialized, including reporters and sources. Diversity in media reporting capac- ity is generally found to relate to diversity in the surrounding community, as Lippmann (1922) has suggested and other investigations have documented (Olien, Donohue, & Tichenor, 1968; Ball-Rokeach & DeFleur, 1976; Chaffee & Wilson, 1977). Even in the small community press there are circumstances that may lead to a certain readiness to deal with conflict (Lee, 19831, such as when a local controversy has already been reported by statewide media.

Under conditions of greater social differentiation, however, conflict report-

Guard Dog Versus Media in a Power Oligarchy

A number of observers appear to treat media as neither independent watch- dogs nor lapdogs, but as part of a power structure oligarchy. Basic elements of this view include (a) media integration into the power structure, but on a more or less equal basis with other powers, and (b) regular interactions that allow the powerful media either to directly oppose other powers in the political and economic realms or to team up with them to attain certain ends. These ingredi- ents are suggested by several writers (Cohen, 1963; Hilgartner & Bosk, 1988; Patterson, 1980), and by transaction theorists (Wolfsfeld, 1991).

Gitlin’s (1980) conception of hegemonic media appears to subscribe to this view, as does Gans (1979), who regards journalists, sources, and audiences as engaging in a tug-of-war over information. Similar views are expressed by Abramson (19901, Arterton (19841, Cater (1959), and Linsky (1986) who view the press as a fourth branch of government able to rally national support for progressive governmental policies. Spragens (1979) finds the media and the public able to team up to oppose corruption in national government.

Some of the more recent agenda-setting research appears to take this oligar- chical perspective. In response to earlier critiques of the 1980 election study, an analysis of the “web of media-political institutions” that govern media atten- tion is provided by Semetko, Blumler, Gurevitch, and Weaver (1991, p. 3). This work modifies the kingmaker view of agenda-setting, but nevertheless holds to a view of media as powerful co-actors.

The guard dog perspective, by contrast, does not regard media as equal co- actors, but as primarily dependent on the dominant powers. What may seem to

121

Journal of Communication, Spring 1995

be a tug-of-war is, from the guard dog perspective, primarily a result of report- ing and reflecting the conflicting views among divided political or economic bodies.

Several other observers who seem to subscribe to the power oligarchy view see the media as constituting a powerful elite (Coyne, 1972; Dye, 1986; Novak, 1974) that interacts with other elites. Thompson (1985) finds media power to be so strong that only unusually strong institutions and leaders can counter it. Dunn (1969) concluded that the press has substantial impact on policy formula- tion, especially when policy is new or changing and the policy makers are inex- perienced.

In elaborating on the role of media as powerful co-actors, Wolfsfeld (1991) has contended that the perspective of structural control over media underesti- mates the power within the media agency itself. His basic argument is that if political and economic powers are so dominant over media, then there should be far fewer complaints about the media from elites, such as U.S. presidents.

Wolfsfeld’s reasoning raises a question: Is hostility between the media and political or economic powers testimony to media independence from these powers? If the press can mercilessly hound a well-known senator or presiden- tial candidate, or a president in difficulty over dealing with Iran or the problem of gays in the military, then how can one then argue that the media are con- trolled by such powers?

view is highlighted by the difference in interpretation of this official hostility toward media. Wolfsfeld’s argument might be reasonable in a totalitarian sys- tem. In a western society, however, this occasional tendency to turn on one of the masters in a pluralistic power structure and yet protect their house is funda- mental to the guard dog conception of media and to the conflict perspective on which it is based. The guard dog perspective holds that media are not a sep- arate or equal power, but that they provide a means by which the power oligar- chy is maintained.

As suggested above, hostility toward media in homogeneous structures is rare. In the more pluralistic structures of urban communities or national politi- cal agencies, the power structure is also the normal beneficiary of media report- ing, but at a price: the guard dog’s conditioning may lead to raising the alarm during internal disputes as well as in the face of clear external threats. The power territory, that is, the structure, is what is being protected, even though specific individual role incumbents may be threatened. In the Minnesota legisla- ture in 1993, allegations of misuse of telephone privileges led to heavy media reporting of what was dubbed a phonegate scandal. The outcome was forced resignation from the positions of house speaker and majority leader by two key Democrats.* The power relationship, however, was not affected in that the legis-

The difference between Wolfsfelds view of media power and the guard dog

The allegations were primarily from the opposition political party in statements quoted in the media. The dismissed speaker of the house, whose adolescent son had passed a legislative account number along to others who made $90,000 worth of unauthorized phone calls, viewed the episode as primarily a media attack on him. His hostility to media reached epic proportions.

122

Journal of Communication, Spring 1995

lature continued to be a contest between Republicans and Democrats, and the Democrats simply replaced the two discredited legislators with other party members. This is not to say that structures never change, but that such change is rare, whereas resolution of conflicts more frequently entails replacement of individuals.

The press, and the media as a whole, are on guard against all intruders so long as the authorities are acting in unison and the power relationships among them are stable. When challenges reach the point of power destabilization, there may be outright confusion about who is running things. A guard dog’s conditioning is such that in this confusion, an alarm is raised in the interest of restoring stability.

When powerful forces challenge a leader displaying threatening behavior, the media will report this challenge in a way that is often regarded as a media attack. Indeed, the reporting may portray the subjects of this reporting as rene- gades or villains, even if they are part of the power structure. President Bill Clinton’s BTU tax proposal received so much media attention because it was a threat to powerful political and commercial interest groups. More generally, any executive leader will be subjected to heavy critical reporting and media commentary when that leader’s positions or actions are seen as threatening within the power structure.

When a controversy continues, part of its resolution may well include dis- placement of blame for the conflict on the media. A U.S. president under criti- cism often says the media are out of control, and/or “blowing the issue out of proportion.” Such assaults on the media may be successful. Hofstadter (1955) pointed out that the journalistic muckraking of 90 years ago, though exposing corruption, was primarily expressing the views of a powerful movement, that is, political progressivism. This era seems to have faded, starting in 1906, when then-President Theodore Roosevelt gave such reporting the pejorative muckrak- ing label and business interests launched a concerted campaign to achieve more favorable coverage (Reynolds, 1979). When political and economic pow- ers do act in concert, they are fully capable of bringing the guard dog media under control.

It should be noted that attacks on the media, such as Spiro Agnew’s attacks on media as “effete snobs” and “nattering nabobs of negativity,” are not neces- sarily acknowledgements of having been damaged by media. Often, the expres- sions are simply instrumental for keeping the media under control, to prevent damage. Where there is a substantial threat to a political position, as in the Watergate crisis, the threat is from other organized political forces, even though the media may become the whipping posts for the role under attack.

In supporting the structure, guard dog media display a tendency to concen- trate on individuals while accepting the structure. In doing so, the media are reinforcing the tendency within the culture to emphasize great men and person- alities rather than individuals as actors in the system who are subject to the influence of social forces and processes. As Lasswell (1948) and Wright (1986) have noted, reinforcing and passing on the cultural heritage is one of the prime functions of media. Political confrontations receiving the most attention

123

Journal of Communication, Spring 19.95

throughout the system and the most media attention are those involving the most highly placed role incumbents, such as presidents, influential members of Congress, and powerful corporate executives.

Marginal individuals and groups, meaning those without power and influ- ence in the system, receive little direct guard dog attention, unless they appear to threaten those at or near the top of the power hierarchies (Gans, 1979). Pro- test groups seeking to capitalize on changing power relationships in the struc- ture may well find the press either ignoring them or reporting them in a negative light. Even when the Vietnam war was an intense political contro- versy, anti-war protests were generally treated as social disorder by the media.

The organization of media and the professional routines of news gathering reinforce this orientation toward power, through selecting sources that are avail- able, efficient and authoritative (Gans, 1979). The press conference, embedded in journalistic routine, is a strategy that succeeds in getting heavy media cover- age primarily for powerful special interest groups and political agencies such as the U S . presidency (Bennet, 1991; Smith, 1990) and the military command in a war situation. Media routines led to coverage of Vietnam that treated mili- tary sources as authoritative (Halberstam, 1972) at least early in the conflict.

War has drawn widespread attention to this tendency to defer to authority, especially where external threats are concerned.) In such conditions of over- whelming general support for political/military action, the deference of media to authority does border on submissiveness. In both Grenada and the Desert Storm coverage, the media acted under the assumption of the U.S. administra- tion that the present power arrangements in the world were threatened and that, consequently, national security was at risk.

Media criticism of military actions, when it occurs, generally reflects either a war that isn’t going well, political controversy over the war, or both. In the case of Vietnam, Hallin (1989) concludes that the “collapse of America’s ‘will”’ (p. 213) to fight as the war dragged on, as well as administration problems with media, resulted largely from political divisions at home. This view is gener- ally consistent with the guard dog perspective, since it locates conflicts, and therefore media agendas, in political interactions that may include wars.

The submission of media to military restrictions during the 1991 Persian Gulf

Implications of the Guard Dog Role

If guard dog is a valid conception of media performance, certain inferences may be stated as hypotheses for further test. A few such inferences are stated here.

3 In the case of the Gulf War, we might note here that before Congress voted to authorize U.S. mili- tary action, public opinion was very sharply divided on whether such action should be taken. In the Minnesota poll, for example, 34% of Minnesota adults supported military action in January, before the vote. In February, after the Congressional vote, 72% supported such action. Typically, there is strong public opinion support for wars after a military initiative is taken. This was true in the Korea, Vietnam and Grenada conflicts.

124

Journal of Communication, Spring 1995

HI: I n any structure, the intensity ofpress reporting and editorializing about a public issue is directly proportional to the degree to which toppowerposi- tions are uncertain as a result of organized challenge.

Maximum uncertainty in the structure occurs when countervailing groups have the capacity to challenge the established powers, and thereby raise a realistic possibility that the power relationships may be altered. A weak group at a Con- gressional hearing gets little attention. Watergate and Irangate, on the other hand, dominated media attention because in each case the legitimacy of the highest position of political executive power in the nation was in question. These were challenges with considerable legitimacy, since they involved com- mittees of the U.S. Congress and such powerful agencies as the Federal Bureau of Investigation (Bernstein, 1974; Colodny, 1991; Mollenhoff, 1981). Although authority issues were at stake in both Watergate and Irangate, it should be noted that these crises contained on the authority roles and not on the system itself.

This power uncertainty hypothesis may be tested directly by considering major policy areas in which there are or are not fundamental divisions between major power groupings. Araby (1989), for example, found that when the L J . 5 executive branch and Congress agreed on a major foreign policy issue, the elite papers’ editorials tended to support that consensus view. When there was a split between Congress and the White House, the papers tended to side with one or the other, but without adopting a third, or innovative, position.

tration of media performance under conditions of power uncertainty. Low regard for the Democratic contenders in the early primaries had a number of bases, among which were long-standing divisions within that party and a cer- tain degree of popular disaffection with Congress, which the Democrats con- trolled. Dislike of President George Bush, though no doubt linked to difficulties in the national economy, was exacerbated by cleavages within the Republican Party over both his tax policy shift and his leadership on such issues as abor- tion (Brookhiser, 1992). This schism led to the Buchanan challenge in the pri- maries and then to an attempt at reconciliation that included Buchanan’s keynote speech to the convention.

These factors provided an opportunity for a third candidacy, an opportunity seized by H. Ross Perot, who possessed immense personal wealth that enabled him to buy his own TV time and facilitated the growth of a third party organiza- tion. His appearance on Lamy King Live contributed to his further visibility as a truly formidable challenger in a politically unstable situation.

After an initial media honeymoon for the Perot campaign, organized chal- lenges to Perot emanated from the established parties and were given promi- nent attention. Perot’s hiring of Hamilton Jordan and Edwin Rollins, prominent strategists from the two existing parties, demonstrated a certain amount of orga- nizational muscle. Whether by design or not, the defections of Jordan and Roll- ins in mid-July indicated chaos in the Perot house, and was followed immediately by the media’s questioning of the credibility of Perot himself.

When Perot returned to the race in time for the fall debates, the general

Media coverage of the 1992 Presidential election campaign provides an illus-

125

Journal of Communication, Spring 19%

theme echoed in the media and held by many observers appeared to be that he would fade quite rapidly, having little or no credibility. On the day of his re- entry, both major parties sent delegations to Houston to negotiate with him for his support, thereby giving Perot renewed credibility.

tified to the impact of a charismatic performance. His televised “infomercials” demonstrated the potential political power of private wealth used to purchase TV time outright. The combined effect of his performance and control of TV time, in the context of Republican Party disarray, contributed to his receiving some 21% of the popular vote.

age that was often critical and probing, as when a reporter pressed him for details about the allegations of dirty tricks against him and his family. Perot learned what many populist politicians have found historically-that guard dog media are fair weather friends for those with marginal positions of power who seek to challenge the existing structure. His decision then to buy his own media time was in keeping with a tradition of marginal groups, such as when a farm movement started its own daily newspaper (Anderson, 1991; Morlan, 1955).

The talk show appearances of the candidates during the 1992 campaign were certainly innovative. The popularity of these appearances with both poli- ticians and audiences has been attributed to their bypassing of reporters. What the talk shows represent is not so much a bypassing as a shift in information control. The talk show host functions much as a master of ceremonies at a pub- lic speech, fielding questions with some discretion. That is, some kind of screening of phone calls typically occurs; for example, a talk show host may cut a question off at any point if he or she wishes. At the same time, the over- all result is to give the politician-source greater control over the exchange. The politician can give unedited responses to questions that are less likely to come from the other party’s perspective, since the reporter is not there acting as a sur- rogate for that other perspective.

His later strong showing in the debates, unanticipated by most observers, tes-

Perot’s ability to maintain that strong a following was in spite of media cover-

H2: Media are more likely to report attacks on individuals in power roles than attacks on power structures.

The Great Person orientation presents a duality and, therefore, a dilemma. Great persons in the system are praised when they enjoy consensus and are subject to attack when the power structure is split. If the general welfare of the system is threatened, the welfare of the individual is at risk and there is little question that in that case the individual will be sacrificed. This hypothesis may be tested through content analysis of coverage of such events as scandals sur- rounding insider trading on Wall Street, Congressional hearings, and major investigations conducted by governmental agencies.

This hypothesis is also relevant to what has been called negative campaign- ing in presidential and other political election campaigns. The attacks on indi- vidual character, such as the reports of the alcoholism problems of Vice

126

Journal of Communication, Spring 1995

Presidential candidate Eagleton in 1972, may be seen as deliberate attempts of political organizations to exploit the conditioned tendencies of media to focus on personalities and, thereby, increase the intensity of coverage. That is, the potential direct effect of the negative ads on the public (Faber, Tims, & Schmitt, 1989) is only part of their impact. The ads may themselves stimulate media attention to the issues they raise.

H3: Guard dog media reporting is less intense when the strategies ofpowerful actors are confined to the traditional rules of political conflict.

It is often noted that coverage of political campaigns has a ritualistic flavor. A primary campaign is predictable in length of time, occurrence of elections, poll reporting, and types of public events. Media coverage by reporters on the bus may be likened to ritualistic barking by a guard dog at many potential occu- pants of the establishment who come and go but are no threat to the power establishment.

Content analyses of media coverage of presidential primary and election peri- ods might provide a test of this hypothesis. The traditional rules of political con- duct in the United States have all but ruled out third party attempts at the presidency. None has ever won the U.S. presidency, attesting to the endurance of the central power structure. There are, however, variations between election years in the media attention given such third party challenges as Theodore Koosevelt’s “Bullmoose” campaign of 191 2 and George Wallace’s American Independent Party of 1968. The Perot insurgency of 1992 is in the same cate- gory, but without designation of a party organization.

One would expect that the amount and direction of media coverage of these various third party attempts would be a joint function of: (a) organizational dis- array within the existing parties and (b) the degree to which the third party can- didate is able to mobilize human and financial resources and thereby establish a presence in primaries or polls.

H4: Media coverage tends to be evaluated as more favorable among groups and organizations occupying more established and dominant powerposi- tions, compared with groups and organizations having less established power.

This hypothesis derives from the point that conflictual relationships between the media and power sources are relatively rare. For the most part, guard dogs find approval from their masters, the claims of official hostility toward media to the contrary. In a Minnesota study, 21 of 29 elected and appointed officials in two communities regarded media coverage as helpful and not harmful, com- pared with only one out of six of leaders of local citizen groups. In another case, 7 of 11 state and federal agricultural officials and established farm group leaders saw media coverage of the American Agricultural Movement as favor- able, compared with only 6 out of 28 members of the AAM itself, which at the time was a newly emerging protest group (Donohue, Tichenor, & Olien, 1984).

Similar findings have emerged from other studies. Linsky (1986) noted that

127

Journal of Communication, Spring 1995

more than 70% of a sample of senior federal officials regarded their relation- ships with the press as more friendly than not. Such evaluation would be expected to reflect the tone of coverage. Nacos (1986), in a longitudinal study of domestic political crises, found that in 66 out of 72 situations, coverage by elite newspapers was more favorable than critical of U.S. presidents.

This is not to deny that established leaders may express negative views of media. In Gunther’s recent analysis, (1992) individuals saw more bias in cover- age of their own groups than in coverage of other groups. Attitudes toward cov- erage of one’s group, however, is not the same as attitudes toward one’s own media interactions. In one study, scientists rated news stories in which they were quoted as “generally accurate” in 94.5% of the cases. When asked about science stories “in general,” 59% said “generally accurate” (Tichenor, Olien, Har- rison, & Donohue, 1970, p. 678). The point here is that this perceived bias would be expected to be greatest among members of groups with relatively lit- tle power in the system. The frequent and harsh criticism of mainstream media by minority group leaders in inner city neighborhoods is a case in point.

H5: Questioning of the guard dog role is more likely in a highly pluralistic structure than in a lesspluralistic one.

In the more homogeneous community, public decision-making operates on more of an informal basis, emphasizing interpersonal communication and a consensus atmosphere. The leadership group tends to be oligarchical and is supported by the local media. Conflict is minimized and decisions are reported largely after they have been made and are treated as consensus outcomes. As a result, the local leadership rarely if ever feels threatened by the media in a more homogeneous structure. Threats to individual actors are more likely to occur in more pluralistic structures, resulting in more questions about media credibility being raised in such communities than in more homogeneous ones.

Several of these hypotheses could also be tested through analyses of Presi- dential primary and convention periods, including the one in 1992. In this cur- rent year, the populist challenge of Ross Perot appears to have received more legitimation in media reporting than most earlier third party challenges. This appears to be largely a result of the widely reported disaffection with the exist- ing two parties and their lack of control early in the primary season over their constituent groups.

tioning of the mass media as an important set of communication agencies in the social system. Media are an important part of the means to an end, but are not ends in and of themselves. Issues are determined not by the media but by those who, by virtue of having attained a position of influence in both public or private sectors, have the capacity to provide informed leadership and estab- lish social agendas. Even in a guard dog role, the media are important to the system. In the competition for resources, media serving as guard dogs for vari- ous interests are a functional part of the balance among the powers that exist, but this does not necessarily mean serving the concerns of the larger public.

The guard dog perspective is offered as a way to better understand the func-

128

Journal of Communication, Spring 1995

References

Ahramson, J. (1990). Four criticisms of press ethics. In J. Lichtenberg, (Ed.), Democracy and the Mass Media (pp. 227-2681. New York: Cambridge University Press.

Altschull, J. H. (1984). Agents ofpower: The role of the news media in human affairs. New York: Longman.

Anderson, J. (1991, August). The role of the Nonpartisan Leader during the Nonpartisan League organization years, 1915-1916. Paper presented to the Assn. for Education in Journalism and Mass Communication, Boston, MA.

Araby, 0. (1989). Thepress and foreign policy: A comparative study of the role of the elitepress in US. foreign policies in the Middle East. Unpublished doctoral dissertation. University of Minne- sota-Minneapolis.

Arterton, F. C. (1984). Media politics: The news strategies ofpresidential campaigns. Lexington MA: Lexington Books.

Bagdikian, B. (1983). The media monopoly. Boston: Beacon

Ball-Rokeach, S. J., & DeFleur, M. L. (1976). A dependency model of mass media effects. Communi- cation Research, 3, 3-21.

Bennet, J. (1991, November). The flack pack: How press conferences turn serious journalists into shills. Washington Monthly, 23, 1€&29.

Bernstein, C. (1974). All the President's men. New York: Simon and Schuster.

Blumer, H., (1954). The crowd, the public and the mass. In W. Schramm, (Ed.), Process and effects of mass communication (pp. 363-398). Urbana, IL: University of Illinois Press.

Breed, W. (1958). Mass communication and sociocultural integration. Social Forces, 37, 107-1 16.

Brookhiser, R. (1992, October). Gravedigger of the revolution: Failure of George Bush to appease conservatives. Atlantic, 270, pp. 70-72.

Burstein, P. (1979). Public opinion, demonstrations, and the passage of antidiscrimination legisla- tion. Public Opinion Quarterly, 43, 157-172.

Carlyle, T. (1841). Heroes: Hero-worship and the heroic in history. New York: Charles Scribner and Sons.

Cater, D. (1959). The fourth branch of government. Boston: Houghton Mifflin

Clarke, P. & Fredin, E. (1978). Newspapers, television, and political reasoning. Public Opinion Quar- terly, 42, 143-160.

Chaffee, S. & Wilson, D. (1977). Media rich, media poor: Two studies of diversity in agenda-hold- ing. Journalism Quarterly, 54,466476.

Chang, T. K. (1989)). Access to the news and U.S. foreign policy: The case of China, 1950-1984 Newspaper Research Journal, 10(4), 33-44.

Cohen, B. C. (1963). Thepress and foreign policy. Princeton, NJ: Princeton University Press

Colodny, L. (1991). Silent coup: The removal of apresident. New York: St. Martin's Press

Coser, L. (1956). The functions of social conJict. New York: Macmillan

Coyne, J. R., Jr. (1972). The impudent snobs: Agnew v. the intellectual establishment. New Rochelle, NY: Arlington House.

Curran, J. (1977). Capitalism and control of the press, 1800-1975. In Curran, J., Gurevitch, M., & Wolkacott, J. (Eds.), Mass communication and society (pp. 195-230). Beverly Hills, CA: Sage.

129

Journal of Communication, Spring 1995

Dahrendorf, R. (1959). Class and class conflict in industrial society. Stanford, CA Stanford Univiver- sity Press.

Dennis, E. E. & Merrill, J. C. (1991). Media debates: Issues in mass communication. New York: Long- man.

Donahew, L. A. (1967). Newspaper gatekeepers and forces in the news channel. Public Opinion Quarterly, 31, 6168.

Donohue, G. A,, Tichenor, P. J., & Olien, C. N. (1984). Media evaluations and group power. In Amo, A. & Dissanayake, W. (Eds.), The news media in national and international conflict (pp 203-2151, Boulder, CO: Westview.

Donohue, G. A,, Olien, C. N., & Tichenor, P. J. (1985). Reporting conflict by pluralism, newspaper type and ownership. Journalism Quarterly, 62, 489-499.

Dreier, P. (1983). The corporate complaint against the media. Quill, 72(11), 17-29

Dunn, D. D. (1969). Public officials and thepress. Menlo Park, CA: Addison-Wesley

Dye, T. R. (1986). Americanpolitics in the media age. Monterey, CA: BrooksKole

Edelstein, A. & Schulz, J. B. (1963). The weekly newspaper’s leadership role as seen by community leaders. Journalism Quarterly, 40, 565-574.

Erskine, H. (1972). The polls: Gun control. Public Opinion Quarterly, 36, 455469

Faber, R., Tims, A,, & Schmitt, K. (1989). Accentuate the negative: The impact ofpolitical appeals on

Gans, H. J. (1979). Deciding what’s news: A study of CBS evening news, NBC nightly news, News-

voting intent. University of Minnesota, School of Journalism and Mass Communication.

week and Time. New York: Pantheon.

Gitlin, T. (1980). The whole world is watching: Mass media in the making and unmaking of the new

Golding, P. (1986). Communicating politics: Mass communications and thepoliticalprocess. New

left. Berkeley, CA: University of California Press.

York: Homes & Meier.

Green, J. C. & Guth, J. L. (1989). The missing link: Political activists and support for school prayer. Public Opinion Quarterly, 53, 41-57.

Gunther, A. C., (1992). Biased press or biased public? Public Opinion Quarterly, 56, 147-167

Gurevitch, M. & Blumler, J. G. (1977). Mass media and politics: A model for the analysis of political communication. In Curran, J., Gurevitch, M., & Wollacott, J. (Eds.), Mass Communication and Society (pp, 27C-290). Beverly Hills, CA Sage.

Hage, G. S., Dennis, E. E., Ismach, A. H., & Hartgen, S. (1976). New strategies forpublic affairs reporting. Englewood Cliffs, NJ: Prentice-Hall.

Hdlberstam, D. (1972). The best and the brightest. Greenwich, CT: Fawcett.

Hallin, D. C. (1989). The “uncensored war’:. The media and Vietnam. Berkeley, CA University of California Press.

Herman, E. S. & Chomsky, N. (1988). Manufacturing consent: Thepolitical economy of the mass media. New York: Pantheon.

Herrera, C. L., Herrera, R., & Smith, E. R. (1992). Public opinion and Congressional representation. Public Opinion Quarterly, 56, 185-205.

Hilgartner, S. & Bosk, C. (1988). The rise and fall of social problems: A public arenas model. Ameri- can Journal of Sociology, 94? 53-78.

Hofstadter, R. (1955). The age ofreform. New York: Alfred A. Knopf.

130

Journal of Communication, Spring 1995

Hohenberg, J. (1973). Freepress, freepeople: m e best cause. New York: Free Press.

Hutchins, R. M. (1947). A free and responsiblepress: A general report on mass communication. Chi-

Key, V. (1961). Public opinion and American democracy New York: Knopf

Lasswell, H. D. (1948). The structure and function of communication in society In L. Bryson (Ed.), The communication of ideas (pp. 37-51). New York: Institute for Religious and Social Studies.

Lee, P. N. (1983). North Dakota weekly newspapers: An assessment of social cohesion and adoption of new technologies. Unpublished master’s project paper. University of Minnesota, Minneapolis.

Linsky, M. (1986). How thepress affects federalpolicymaking: Six case studies. New York: Norton.

Lippmann, W. (1961). Drift and mastey. Englewood Cliffs, NJ: Prentice-Hall. (Original work pub-

cago: Commission on Freedom of the Press and Chicago, University of Chicago Press.

lished 1914)

Lippmann, W. (1922). Public Opinion. New York: Macmillan

McCombs, M. (1992). Explorers and surveyors: Expanding strategies for agenda-setting research Journalism Quarterly, 0,813-824.

tion, 43(2), 58-67, McCombs, M. & Shaw, D. (1993). The evolution of agenda-setting research. Journal of Comrnunica-

Merrill, J. C. & Lowenstein, R. L. (1971). Media, messages and men. New York: McKay.

Mills, C. W. (1956). Thepower elite. New York: Oxford University Press.

Mollenhoff, C. R. (1981). Investigative reporting: From courthouse to white House. New York: Mac- millan.

Molotch, H. & Lester, M. (1975). Accidental news: The great oil spill as local occurrence and national event. American Journal of Sociology, 81, 235-260.

Morlan, R. (1955). Politicalprairiefire. Minneapolis: University of Minnesota Press.

Morley, D. (1976). Industrial conflict and the mass media. The Sociological Review, (2412, 245-268.

Nacos, B. L. (1990). Thepress, presidents, and crises. New York: Columbia University Press.

Novak, R. D. (1974). The gap between the liberal voters and the masses. In H. M. Clor (Ed.), The mass media and modern democracy (pp. 1-14). Chicago: Rand McNally.

Olien, C. N., Donohue, G. A., & Tichenor, P. J. (1968). The community editor’s power and the reporting of conflict. Journalism Quarterly, 47, 472-478.

Page, B., Shapiro, R., Gronke, P., & Rosenberg, R. (1984). Constituency, party, and representation in Congress. Public Opinion Quarterly, 48, 741-756.

ity. Public Opinion Quarterly, 35, 808-892. Paletz, D. L., Reichert, P., & McIntyre, B. (1971). How the media support local government author-

Patterson, T. E. (1980). The mass media election: How Americans choose tbeirpresident. New York:

Reynolds, R. D., Jr. (1979). The 1906 campaign to sway muckraking periodicals. Journalism Quar-

Praeger.

terly, 56, 513-520.

nal of Communication, 43(2) 68-84, Rogers, E. M., Dearing, J. W., & Bregman, D. (1993). The anatomy of agenda-setting research. Jour-

Rossell, Deac, (1978). The fourth estate and the seventh art. In B. Rubin (Ed.), Questioning media ethics (pp. 232-2821, New York: Praeger.

Schrarnm, W. (1957). Responsibility in mass communication. New York: Harper.

131

Journal of Communication, Spring 1995

Schulte, H. (1981). Reportingpublic affairs. New York: Macmillan.

Semetko, H. A,, Blumler, J. G., Gurevitch, M., & Weaver, D. H. (1991). The formation of campaign agendas: A comparative analysis ofparty and media roles in recent American and British elec- tions. Hillsdale, NJ: Erlbaum.

Shepherd, G. (1981). Selectivity of sources: Reporting the marijuana controversy. Journal of Commu- nication, 31(2), 129-137.

Sigal, L. V. (1973). Reporters and officials: The organization andpolitics of newsmaking. Lexington, MA: D. C. Heath.

Small, W. J. (1972). Politicalpower and thepress. New York: Norton.

Smith, C. D. (1990). Presidentialpress conferences: A critical approach. New York: Praeger

Spragens, W. C. (1979). Thepresidency and the mass media in the age of television. Washington, DC: University Press of America.

New York: Lanham. Thompson, K. W. (1985). The media: The credibility of institutions, policies and leadership (Vol. 5).

Times-Mirror Corporation (1986). Is the watch-dog really a lap-dog? [Advertisement]. Columbia Jour-

Tichenor, P. J., Donohue, G. A,, & Olien, C. N. (1980). Community conflict and thepress. Beverly

nalism Review, 74(2), 11.

Hills, CA: Sage.

Tichenor, P. J., Olien, C. N., Harrison, A,, & Donohue, G. A. (1970). Mass communication systems and communication accuracy in science news reporting. Journalism Quarterly, 47, 673-683.

Vaughn, S. & Evensen, B., (1991). Democracy's guardians: Hollywood portrait of reporters, 1930- 1945. Journalism Quarterly, 68, 819-838.

election. New York: Praeger. Weaver, D., Graber, D., McCombs, M., & Eyall, C. (1981). Media agenda-setting in apresidential

Wolfsfeld, G. (1991). Media, protest, and political violence. Journalism Monographs, 127.

Wright, C. R. (1986). Mass Communication: A sociologicalperspective (3rd ed.). New York: Praeger.

132