Abstract
Many journalists and other observers remember the 1960s as a watershed moment in American journalism. Do they remember correctly? This essay reviews relevant empirical studies on how US newspapers have changed since the 1950s. There is strong existing evidence that journalists have come to present themselves as more aggressive, that news stories have grown longer, and that journalists are less willing to have politicians and other government officials frame stories and more likely to advance analysis and context on their own. Based on content analysis of the New York Times, Washington Post, and Milwaukee Journal Sentinel, this study finds that the growth in ‘contextual reporting’ has been enormous – from under 10 percent in all three newspapers in 1955 to about 40 percent in 2003; ‘conventional’ news stories on the front page declined from 80–90 percent in all three papers to about 50 percent in all three papers in the same period. What this study calls ‘contextual reporting’ has not been widely recognized (unlike, say, investigative reporting) as a distinctive news genre or news style and this article urges that it receive more attention.
Keywords
As virtually all accounts by journalists and historians attest, news coverage of government, politics, and society opened up in the 1960s and 1970s. This was the joint product of three developments: the culture of journalism changed and journalists asserted themselves more aggressively than before; institutions of government changed, becoming less secretive and more attuned to the news media; and the very concept of ‘covering politics’ was redefined as the federal government expanded its reach (in civil rights, economic regulation, environmental responsibility, and social welfare programs like food stamps, Medicare, and Medicaid) and as the women’s movement emerged and proclaimed that ‘the personal is political’. The story of a transformed journalism has been told many times before, but it has generally failed to specify what this transformation looked like in the pages of the newspapers. Much attention has focused on the growth of investigative reporting, a vital but small part of journalism’s transformation. In this article, we focus on the quantitatively most significant change in newspaper journalism between the 1950s and the early 2000s – the rise of what, for lack of a better term, we will call contextual reporting. More than other concurrent changes, this one altered the front page, but this has received little academic or popular attention – little enough that standard accounts have not even come to agreement on a name for it.
Journalism’s coming of age in the 1960s and 1970s
The decade from the mid-sixties to the mid-seventies was the end of innocence for many Americans. This certainly included journalists. As many of them have recalled, reporting grew less deferential to politicians and more insistent and probing from the mid-1960s on. Political scientist Michael Robinson, interviewing journalists, members of Congress, and congressional staffers between 1977 and 1980, found near unanimity that a more aggressive and critical news media was the biggest change in Congress–media relations in their experience. ‘Ask anybody on Capitol Hill about the most basic change in the relationship between Congress and the media since 1960,’ he wrote, ‘and the response is practically catechistic – the media have become harder, tougher, more cynical’ (Robinson, 1981: 55).
Media coverage of Congress in the 1950s and into the 1960s was, as one contemporary called it, ‘overcooperative’ (Matthews, 1960: 207). One reporter on Capitol Hill said (in 1956) that covering the Senate was ‘a little like being a war correspondent; you really become a part of the outfit you are covering’ (Matthews, 1960: 214). In the House during the 1940s and 1950s, cooperation was orchestrated by a powerful Speaker, Sam Rayburn. For Rayburn, politics was about what happened in the Congress, not what outsiders said about it. Rayburn was unwilling to participate on television’s Sunday news interview programs and he banned cameras on the House floor as well as the recording of House committee meetings. At the same time, the Speaker invited an inner circle of trusted reporters to off-the-record sessions of drinking and discussion at the end of the working day and his daily five-minute press conferences were almost totally controlled by these insiders, who protected him from any difficult questions (Foote, 1998).
The culture of the press was cooperative, even complaisant. ‘Until the mid-1960s,’ as historian Julian Zelizer writes, ‘the press was generally respectful of the political establishment’ (2007: 230). The decline of this respect helped bring more attention to political scandal. Scandal reporting is frequently decried as a lowering of the standards of the press from serious and fair-minded coverage of issues to a frivolous and sensational focus on political sideshows. In this view, scandal reporting encourages citizens’ alienation from a politics portrayed as terminally tawdry. But scandal reporting is also a symptom of a system that had become more democratic. As governing became more public (the Legislative Reorganization Act of 1970 was among the milestones in ‘sunshine’ legislation), politicians and government officers have been more often held accountable, and the character of democracy shifted from one in which voters normally could express disapproval of government incumbents only on election day to one in which, in Zelizer’s words, ‘the nation would no longer have to wait until an election to punish government officials, nor would it have to depend on politicians to decide when an investigation was needed’ (2007: 236). To some extent, the proliferation of scandals was made possible not only by the availability of new information (for instance, campaign finance disclosure after the reform acts of 1971 and 1974) but by the growing acceptance of values promoted by the women’s movement that blurred the line between public and private behavior.
But are recollections of a changing political culture in Washington and around the country confirmed by close analysis of news content? There is not very much systematic examination of changing news forms and styles during the past seven decades, but what there is generally confirms what journalists and politicians themselves recall. Still, the data we have gathered from existing studies and from our own content analysis, described below, depart from memory in one important respect. The decline in deference to government officials, as expected, led to an increase in what is called investigative reporting, but that increase was quantitatively modest. Meanwhile, there was a stunning growth in what we will call ‘contextual reporting’. As Kathy Roberts Forde has observed, there is no standard terminology for this kind of journalism. It has been called interpretative reporting, depth reporting, long-form journalism, explanatory reporting, and analytical reporting (Forde, 2007: 230). In his extensive interviewing of Washington journalists in the late 1970s, Stephen Hess called it ‘social science journalism’, a mode of reporting with ‘the accent on greater interpretation’ and a clear intention of focusing on causes, not on events as such (1981: 57). Although this category is, in quantitative terms, easily the most important change in reporting in the past half century, it is a form of journalism with no settled name and no hallowed, or even standardized, place in journalism’s understanding of its own recent past.
The following five propositions summarize the evidence others have gathered that contribute to identifying the family of changes that represents a shift in the culture of journalism: news has grown more critical of established power; journalists have come to present themselves publicly as more aggressive; news stories have grown longer (and presumably deeper); news stories have grown less government and electoral politics centered; and news has grown more contextual. We summarize the available evidence on these points and follow that with our own data on the fifth point – the growth of contextual reporting.
Changes in newspaper news, 1950s to 2000s
News has grown more critical of established power
A study of campaign coverage of presidential elections by political scientist Thomas Patterson found that news coverage (in weekly news magazines) grew increasingly negative from 1960 to 1992. In 1960, 75 percent of evaluative references to candidates John Kennedy and Richard Nixon were positive; by 1992, only 40 percent of evaluative references praised Bill Clinton or George Bush (1993: 20). Not only were evaluations more negative than they had been but campaign coverage grew implicitly more cynical in focusing on the ‘game’ of politics rather than the policies proposed by candidates. Over the same 1960–92 period, as Patterson found in another study, news accounts in the New York Times paid relatively less attention to what candidates said in speeches and relatively more attention to the political strategies behind the speeches. In his terms, this was a move from a ‘policy’ framework for reporting to a ‘game’ framework (1993: 11–12, 68–77). It shifted from what politicians said to the political context in which they said it, implicitly or explicitly contending that the strategic or political context helped explain the politicians’ policy pronouncements and other statements.
Looking at 10 mainstream metropolitan dailies from 1963–64 and 1998–99 (sampling two weeks in each period) from different regions around the country, Carl Sessions Stepp wrote, ‘To read 1963 newspapers is to re-enter a pre-Watergate, pre-Vietnam, pre-Dealey Plaza world. It is to roll back a gigantic cultural loss of idealism.’ According to Stepp, newspapers in this earlier period ‘seem naively trusting of government, shamelessly boosterish, unembarrassedly hokey and obliging’. He was surprised to find stories ‘often not attributed at all, simply passing along an unquestioned, quasi-official sense of things. The world view seemed white, male, middle-aged and middle class, a comfortable and confident Optimist Club bonhomie’ (1999: 65). This was completely different from what he found in his 1998–99 sample. Journalists celebrate their tradition of savviness, critical judgment, and an instinct for the soft underbelly of politicians, but Stepp’s analysis simply finds little evidence of any of these features in the content of 1963–64 newspapers.
Journalists have come to present themselves publicly as more aggressive
In a rich series of research papers, sociolinguist Steven Clayman and his colleagues analyzed the questions reporters have asked in presidential press conferences from 1953 through 2000. They found significant increases in ‘initiative’ (prefacing a question with statements to construct a particular context, asking multiple questions within a single turn, or asking a follow-up question), in ‘assertiveness’ (inviting a particular answer – ‘isn’t it true that …? or ‘don’t you think that …?’); and in ‘adversarialness’ (‘Mr President, Senator So-and-So has criticized your Policy X as disastrous for the economy, national defense, and American morals – how do you respond?’). There was a notable rise on all of their measures of aggressiveness in 1969 and at no point after 1969 has the heightened level of aggressive questioning returned to the deferential questioning style that prevailed during the Eisenhower, Kennedy, and Johnson administrations. The scholars find the only plausible explanation for this is a ‘normative shift’ in the practice of journalism. No other contextual variables – the party of the incumbent President, the state of the national economy, the extent of divided government – explain the persistence of aggressive questions that attempt to hold the President accountable (Clayman et al., 2006, 2010).
Clayman and his colleagues examined the transcripts of presidential press conferences, not newspaper and television stories based on them. So what they document is change in how reporters have presented themselves. That Washington journalists came to style themselves as tough and assertive, people able to stand their ground face-to-face with the President of the United States, is important in itself. They did not then necessarily produce stories that mirrored their assertive questioning, but it is reasonable to infer that to some degree they did.
News stories have grown longer
One well-documented study, by Kevin Barnhurst and Diana Mutz, shows that newspaper stories have become longer over time. Sampling the New York Times, the Chicago Tribune, and the (Portland) Oregonian every 20th year from 1894 through 1994, Barnhurst and Mutz find a consistently increasing mean length of news stories in all three papers, in all three categories of stories that they examined (accidents, crimes, and job-related stories) across the whole time-span of their study – 1894 to 1994. The three papers showed little change from 1914 to 1934; the Oregonian shows a notable increase in length by 1954 and all three papers – the Times especially – show growth between 1954 and 1974. Stories in the Times and the Oregonian continued to lengthen in 1994, although the increases are modest; the Tribune story length decreased between 1974 and 1994 but remained higher than in any of the years measured from 1894 through 1954 (Barnhurst and Mutz, 1997: 32; see also Barnhurst, 1991: 110).
These results are consistent with Stepp’s findings. Stepp found a large reduction between 1964 and 1999 in the number of very short stories and an increase in the number of very long stories. Stories less than six inches long declined from 36 in the front section of the 10 papers he examined to 13; inside the front section (excluding page one), stories longer than 20 inches increased from one to three. On the front page, where most of the very long stories began, average story length increased from nine inches in 1964 to 20 inches in 1999 (1999: 75).
Barnhurst and Mutz do not certify that the longer stories of 1974 and 1994 offer ‘better’ journalism than the shorter stories of 1954 and earlier, but it is hard not to believe that, in general, this is so. Stepp acknowledges that the 1999 papers struck him as ‘less flavorful, less surprising, and – distressingly – less imbued with a distinctive sense of place’ than those of 1964. Nevertheless, he judges that the 1999 papers were ‘by almost any measure, far superior to their 1960s counterparts’. They were ‘better written, better looking, better organized, more responsible, less sensational, less sexist and racist, and more informative and public-spirited’ (1999: 62).
News stories have grown less government and electoral politics centered
Non-political stories – that is, stories that do not focus specifically on governmental actors or electoral campaigns – have appeared more frequently on the front page over time. In a study by Stephen Hess, looking at a single week’s New York Times’ coverage, government or political stories declined from 84 percent in 1965, to 73 percent in 1975, to 63 percent in 1985, and to 55 percent in 1992 (1994: 148). In our view, this should be understood not so much as a ‘decline’ of political coverage but as a greatly expanded understanding of what counts as a matter of general public significance and political relevance. Stories about health, science, business, the arts, social trends, and profiles of individuals whose experiences related to current political or social concerns but who are not themselves political actors so much as beneficiaries or victims of public policies, all appear more often than in the past. Newspapers have increasingly sought to provide the stories behind and beyond the old ‘the public agenda is what leading political figures say it is’; they even try to get at topics that might next week, next month, or next year explode into issues that will demand the attention of politicians and governments.
News has grown more contextual
Thomas Patterson’s research found that in 1960 more than 90 percent of front-page stories concerning electoral campaigns in the New York Times were descriptive – but less than 20 percent by 1992 (1993: 82–83). Patterson suggests that the growth of a more interpretive style in the newspapers was the adoption of ‘the television model’ (1993: 81). This suggests an explanation for change – that newspapers copied television to keep up with it, but that does not strike us as sufficient. Patterson may be closer to the truth when he characterizes a changed attitude among journalists: ‘Journalists had been silent skeptics; they became vocal cynics’ (1993: 79). This is consistent with the conclusion of Clayman et al. (2010) that a cultural change or ‘normative shift’ among journalists is the best explanation of the growing assertiveness of reporters after 1968.
Barnhurst (2011: 100–101, 114) finds a consistent growth from 1914 through 2005 in the percentage of front-page stories that refer to either the past or the future (usually the past) rather than to the temporally immediate context (hours or days) of the event the stories focus on. The rate of growth is highest in the 1994–2005 period, but there is an increase throughout. Barnhurst plausibly sees references to other time periods as an index of the tendency to provide a story with more context.
Also consistent with this conclusion is research on the shrinking of the sound bite in television news. In national network coverage of presidential elections, the average length of time a candidate spoke uninterrupted on camera was 43 seconds in 1968; by 1988, it was nine seconds. Does this mean that television news has grown more trivial? That is not the conclusion of Daniel Hallin’s careful research on the topic. His view is that TV news had grown more ‘mediated’, that is, that journalists intervened with growing frequency in order to provide a compact and dramatic story – and in order not to let the candidates control the story. What this meant for the overall content of TV news, Hallin found, was an increase in ‘horse race’ coverage, a measure of the growing prominence of a ‘game’ or ‘strategy’ orientation in the news. This seems to confirm everyone’s worst fears, but Hallin also found an increase in ‘issues’ coverage, showing that television was performing exactly as media critics wished. How can both kinds of coverage increase at the same time? Doesn’t one grow only at the expense of the other? Hallin’s answer is that with the shorter sound bites, reflecting an increasingly interventionist stance of TV journalists, TV news offered a ‘more highly structured, thematic’ story. In the new TV news, there is less wasted motion, less silence, more rapid-fire editing – and thus more time devoted to both ‘horse race’ and ‘issues’ (1994: 145, 147).
We decided to tackle head-on trends in contextual news stories compared to trends in conventional news stories in newspapers from the 1950s to the 2000s. The next section presents our work, which does indeed confirm an increase – a remarkably large increase – in contextual news.
Are the trends toward a more critical or skeptical journalism the same as trends toward a more active, aggressive, and investigative journalism? Are the performances of journalists at White House press conferences evidence that journalists have internalized the values of a more assertive journalism or only that they recognize its market-pleasing or career-making value? Are any of these measures also evidence that a more analytical or contextual journalism has gained ground or are all of these separate trends? Reasonable people can differ on these questions, but our own sense is that all of these trends demonstrate a consistent movement away from the cautious, formulaic, cut-and-dried conventional journalism practiced in the 1950s and early 1960s.
Changing types of front-page stories, 1955–2003
Over the second half of the 20th century, did news become less about documenting the important events of a given day and more about providing context? Has there been a decrease in conventional, ‘just the facts’ types of stories and an increase in contextual reporting?
In a word, yes. We undertook a content analysis of three newspapers: the New York Times, the Washington Post, and the Milwaukee Journal Sentinel. We selected the Times and the Post because of their central importance as national leaders of American journalism from the late 1960s to the present; we chose the Journal Sentinel as a well-regarded regional daily to stand in for the many strong papers around the country that dominate news-gathering in a region, papers like the Miami Herald, Boston Globe, St. Louis Post Dispatch, Arizona Republic, and others. We examined articles on the front pages of each newspaper over two constructed weeks (Monday of Week 1, Tuesday of Week 2, Wednesday of Week 3, and so forth) in the years 1955, 1967, 1979, 1991, and 2003. We steered clear of election years so that campaign stories did not overwhelm other news. Based on their content and style, each of the 1891 articles in the sample was assigned to one of five possible categories: conventional, contextual, investigative, social empathy, or ‘other’.
Conventional stories often, though not always, inform the public about the official activities of government. This category includes stories about lawmaking and politics, but also public safety, such as court prosecutions, police crime reports, and responses to fires and natural disasters. A conventional story as we understand it, however, is defined not by its subject matter but by its approach. Three features stand out. First, a conventional story identifies its subjects clearly and promptly. Commonly, these stories answer the ‘who-what-when-where’ questions in the lead paragraph or even the lead sentence. Also, commonly, the stories ignore or only implicitly address the ‘why’ question. They tend to be written in the ‘inverted pyramid’ style, with the most important information coming first.
Second, the conventional story describes activities that have occurred or will occur within 24 hours. (In some cases, the activities may have occurred earlier but were not publicly known until very recently.) One giveaway of a conventional story is a lead paragraph with the word ‘yesterday’ or ‘today’.
Third, conventional stories focus on one-time activities or actions. This includes planned events, such as public meetings, as well as unplanned actions like natural disasters. These activities may not be events in the world, but rather statements about them made by a powerful person, either in public or in speaking with a reporter.
The article ‘Conferees Approve 8.8 Pct. Pay Raise’ from the 4 May 1955 edition of the Washington Post is an example of a conventional story. It meets all three of our criteria in just its lead sentence: ‘Senate-House conferees yesterday agreed on an 8.8 percent average pay raise for the Nation’s 500,000 postal employees, including 15,000 here and nearby.’
Contextual stories tend to focus on the big picture, providing context for other news. If the conventional story is a well-cropped, tightly focused shot, the contextual story uses a wide-angle lens. It is often explanatory in nature, sometimes appearing beside conventional stories to complement the dry, ‘just the facts’ versions of that day’s events. Sometimes, newspapers label contextual stories ‘news analysis’, as if to head off anticipated criticism that these stories mix interpretation with facts. Contextual stories are often written in the present tense, since they describe processes and activities that are ongoing rather than events that have been both initiated and completed in the preceding hours or days. Alternatively, they may be written in the past tense, if their purpose is to give historical context.
Contextual stories are not all alike. They may be explanatory stories that help readers better understand complicated issues. They may be trend stories, using numerical data that show change over time on matters of public interest like high school graduation rates, population growth or unemployment. Trend articles in the later years of our sample often included charts or graphs. There are also descriptive stories that engage the imaginations of readers, transporting them to unfamiliar places. These are not travel pieces – they describe places that are newsworthy but perhaps unfamiliar. An Associated Press story printed in the Milwaukee Journal Sentinel (called just the Milwaukee Journal at the time) in 1967, for example, is based on a reporter’s observations in Communist China: Canton, China – AP – Though one million Red Guards throng Canton’s streets and are in effective control, there are no civil disorders. The opposition to the guards is covert, furtive and virtually underground. Police are still on duty, most of them wearing armbands to show they are in the Red Guards. But traffic in the streets of south China’s biggest city is limited to a handful of trucks, buses and trolley buses.
Descriptive contextual articles are not always about places far from home. A New York Times article from 1991 describes two competing images of Newark, New Jersey: ‘one of gleaming steel and glass towers, the other of 100-year-old railroad shacks and multifamily wood frame houses in neighborhoods with few stores or amenities, not even a movie theater’. Several descriptive contextual stories in our sample also provided a backdrop for military activity that was underway or imminent. Newspapers in 1991 featured many accounts by reporters who described what they saw as they were embedded with the US military in Iraq. There are different ways to offer context; what all contextual stories share is an effort at offering analysis or context that goes beyond the ‘who-what-when-where’ of a recent event.
Although conventional and contextual stories comprised the majority of the articles in our sample, we made note of two other types: investigative and social empathy stories. In investigative stories, the newspaper is clearly playing ‘watchdog’ –investigating corruption or coming to the aid of a person who has been treated unjustly. Investigative stories often require extensive time and research, due in part to resistance from sources who fear the stories will reflect poorly upon them. Reporters may have to compel government sources to provide information via formal requests for public records, such as through the Freedom of Information Act or state open records laws. Other information for investigative stories comes from confidential sources. Reporters often call attention to such methods in the ways they attribute their sources: ‘according to documents obtained by [news organization]’. For the purposes of our coding, articles that referenced efforts like these – obtaining non-public documents, or conducting many or lengthy interviews – were considered to be investigative.
Social empathy stories describe a person or group of people not often covered in news stories. They may answer the question, ‘what does it feel like to be this person?’ Such stories encourage readers to be interested in, have compassion for, or empathize with, the experiences and problems of people who are largely unfamiliar to them. Social empathy stories often use personal experiences to highlight larger social problems – such as the many stories of struggle in New Orleans’ Lower 9th Ward in the aftermath of Hurricane Katrina in 2005. Social empathy stories often begin with anecdotal leads, use many direct quotes from their main sources, and structure narratives around the observations of sources rather than those of a detached observer. Examples of social empathy stories in our sample included a 1979 Milwaukee Journal article about the political struggles of India’s untouchables, and a 2003 article in the Washington Post about the trauma experienced by witnesses to two suicide bombings in Istanbul. Social empathy and investigative stories are specific brands of contextual journalism, distinctive enough and important enough to be counted separately, but they can be added to the sum of contextual news stories to measure the general shift away from conventional reports.
We classified stories that did not fit any of these four categories as ‘other’. These include a front-page editorial (regularly in the lower left corner) that ran daily in the Milwaukee Journal Sentinel into the early 1980s. The newspaper did not label this feature an editorial until 1976, but it clearly dedicated a particular spot on its front page for this purpose. Usually titled ‘Milwaukee’ or ‘On, Wisconsin’, these columns usually contained analysis, but also clearly crossed the line over to opinion. The Washington Post occasionally ran commentaries on its front page in the 1970s. We coded these pieces as ‘other’, too.
The ‘other’ category also included what are sometimes called ‘water cooler’ stories, or ‘kickers’. These stories, which appeared most often in the earlier years of our sample, took the form of vignettes that were amusing or shocking. The aim of these stories was not to inform readers, like conventional stories, to explain things, like contextual stories, to remedy injustices, like investigative stories, or to inspire compassion, like social empathy stories. Water cooler stories were designed to entertain. One example is ‘D.C. Playground Dedication Speech Ends With One Great Big “Waaaaa!”’ from the 25 April 1955 edition of the Washington Post. The story describes a ‘momentary crisis’ when a four-year-old boy who was going to speak about his appreciation for the new playground instead got stage fright and cowered in his mother’s arms. ‘It’s a pity,’ the newspaper quotes the boy’s mother as saying. ‘He practiced so hard.’
Some stories that were likely to be told around a water cooler were categorized as conventional if they fit our three criteria. Take, for example, this Associated Press brief, ‘He’ll Need Interpreter’, which ran in the 13 January 1955 edition of the Milwaukee Journal: Jefferson City, Mo. – (AP) – A thief took a typewriter from the department of modern languages at Lincoln university Wednesday. The machine prints in Czecho-Slovakian.
Clearly, this story was included for its humor, not its newsworthiness. Still, we categorized this story as conventional because it (1) stated its subjects plainly and promptly, (2) recounted something that occurred the previous day, and (3) focused on a one-time activity.
The line between conventional and contextual stories blurs, especially in the later years of our sample, when stories that used an inverted pyramid structure also often included a heavy dose of analysis. Interpretive clauses are embedded within the who-what-when-where statements typical of conventional articles, as if the reporters are interrupting their own thoughts to contextualize the story as they go along. Consider this lead paragraph from ‘Bush Proposes North Korea Security Plan to China’, in the 20 October 2003 edition of the New York Times: BANGKOK, Monday 20 October – President Bush presented President Hu Jintao of China with a new, if still vague, American plan here on Sunday that would provide a five-nation security guarantee to North Korea – but not a formal nonaggression treaty – if the North dismantles all of its nuclear weapons programs. (emphasis added)
The italicized phrases clearly represent the reporter’s attempt to contextualize the security guarantee. By describing it as ‘still vague’, the lead paragraph suggests that the new plan is not very different from the last one the US proposed. And by mentioning that the security guarantee is not a formal nonaggression treaty, the reporter clarifies that the USA is not agreeing to give North Korea what it really wants. How, then, to categorize this story? We chose to count it as a conventional story since its general structure was strictly conventional and the interpretive or contextual elements, though of an analytical bent rarely to be found in conventional news stories of the 1950s and 1960s, are presented as parenthetical rather than central.
Coding is not foolproof, even with the relatively simple, relatively clear – and relatively few – categories we have employed. Even so, we do not see any reason to doubt the general trend lines our data reveal. 1
Results
Figure 1 and Table 1 summarize our findings for the three newspapers in our study combined (see also Appendix for exact percentage figures for each newspaper):
Findings for the three newspapers combined.
Findings for the three newspapers combined.
Once the dominant style of newspaper writing, conventional stories no longer overwhelm the front page. Meanwhile, the number of contextual stories has grown tremendously. The numbers of investigative and social empathy stories have inched upward over the years, but they remain a small fraction of news stories overall. Front-page ‘water cooler’-type stories have disappeared. The shift from conventional articles to contextual articles was most pronounced between 1955 and 1979. The decline in conventional stories continued in subsequent years, but at slower rates (Table 2).
Percentage change in share of total front page articles.
Examining the types of stories as a percentage of all front page stories is important, since the total number of stories on newspaper front pages declined drastically during these years (Table 3). Newspapers by 1991 had less than half as many articles on their front pages as they did in 1955. The average number of articles per page was 13.5 in 1955; by 2003, that number declined to 7.3. Part of the reason for the drop is that articles became longer (Barnhurst and Mutz, 1997; Stepp, 1999). Newspapers also began to devote more space to non-article items, such as photos and promotional ‘teasers’ (boxes or columns with paragraphs aimed at enticing you to read articles inside the newspaper).
Number of stories.
The number of investigative stories was too small to draw any definitive conclusions about their growth over time. More significant is the obvious fact that investigative stories have always been, and continue to be, a rarity. This is not surprising. Investigative work requires considerable time and effort, and sometimes does not even lead to a story – if a tip turns out to be wrong, or if reporters encounter an insurmountable roadblock in uncovering the information they need. Also worth noting is that contextual stories can perform a similar function to investigative stories by shedding light on issues that deserve more public attention, even if there is not a recent event to anchor them. While contextual stories may not require battles over the acquisition of public records, or Watergate-style interactions with confidential sources, they may still uncover information that is available, but unexamined. Take, for instance, this excerpt from ‘Showdown Over Loudoun; With Growth at Issue, Builders’ Cash Fuels Races’, an article from the 29 October 2003 edition of the Washington Post, about politics in northern Virginia’s Loudoun County: Since July 2002, builders and other real estate interests have given candidates running for Loudoun’s Board of Supervisors more than $460,000 – seven times what they dispensed during the same nearly 16-month period before the 1999 election, according to a Washington Post analysis of reported campaign contributions. Among the donors are four of the nation’s 10 biggest home builders.
The number of social empathy stories increased notably between 1967 and 1979 and has remained steady since then. Many social empathy stories are about people during and after war – soldiers and civilians.
We examined whether Hess’s (1994) finding of a decline in political news over a similar time period might explain the shift from conventional to contextual news. If government stories past and present are typically conventional, and non-government stories past and present normally contextual, our findings may just be an unnecessarily complicated way of saying that there is less front-page coverage of government than there used to be. What can be said about this possibility?
We found, as Hess did, a decline in coverage of politics and government in the New York Times. Our results, though, indicate a less severe decrease and, more recently, a leveling-off. Results from the Washington Post and Milwaukee Journal Sentinel are different. Politics and government coverage in the Post increased for a time before returning to its 1955 level in 2003. The Milwaukee Journal Sentinel actually had more politics and government coverage in 2003 than it did in 1955, although the upward trend was hardly consistent (Table 4).
Percentage of front-page politics/government stories.
These results suggest a methodological challenge. Counting the number of politics and government stories that appear on newspaper front pages is not simple. A president’s policy speech certainly qualifies, but what about his more mundane activities, like going on vacation or visiting the hospital for an upset stomach? Supreme Court rulings are appropriately labeled as politics/government, but what about local court cases – what many people would consider ‘crime’ stories? War coverage also obviously involves politics and government, but what about stories that focus on the lives of soldiers at the front or of civilians at home touched by the war? In our attempt to be as consistent as possible, we labeled all of these stories as politics/government. The results reiterate what many scholars have concluded: the vast majority of front-page news coverage has been and continues to be about politics and government. What we have found is that while politics and government continue to dominate news stories, the ways journalists tell these stories have clearly changed from an almost completely conventional style to a mix of conventional and contextual. Even in stories about government and politics, there has been a change from conventional to contextual reporting.
We also looked at the authorship of our sample articles to determine whether the amount of content contributed by media outlets other than the newspaper’s own staff writers might explain the shift from conventional to contextual news. These other media outlets include wire services like the Associated Press, Reuters and United Press International, as well as other newspapers like the Los Angeles Times and New York Daily News. If the preponderance of stories from outside agencies have been written in conventional form, might our results be explained simply by a declining use of outside media?
In fact, although the percentage of front page news stories that originated in other media varied among the newspapers we studied, all of them used less outside content over time (Table 5). And, yes, stories from outside sources have been far more likely to be conventional than in-house stories but only at the Milwaukee Journal Sentinel could outside stories conceivably contribute to explaining the rise of contextual journalism. In the Milwaukee Journal Sentinel, while conventional stories from outside media sources stay between 74 percent and 85 percent across the years in our sample, in-house conventional stories drop from over two-thirds in 1955 and 1967 to well under half in the three subsequent years we examined (Table 6). The decline in the use of wire service stories in our sample contributes to the sharp upturn in contextual stories on the front page, but it is just one contributing factor (and in the New York Times, which used very little outside material even in 1955, it is scarcely a factor at all).
Percentage of front page stories from outside sources.
Percentage of outside and in-house conventional stories at Milwaukee Journal Sentinel.
Conclusion
Sometime around the late 1960s, the tenor of Washington journalism began to change. A growing body of research converges in its portrayal of a shift toward increasingly vigorous and in some respects adversarial treatment of government officials, political candidates, and their policies. (Clayman et al., 2010: 229)
Journalists themselves have placed so much emphasis on Watergate (1972–74) as a turning point that they sometimes forget that (a) the big change in the news culture began in the late 1960s, and (b) the growth of contextual journalism represents a much larger quantitative change than a reallocation of effort to investigative reporting.
Meg Greenfield, a journalist who became an editorial writer at the Washington Post in 1968, and editor of the editorial page in 1979 until her death in 1999, said it most simply in her posthumous memoir: ‘The great change in Washington began in the late 1960s’ (2001: 83). She recalled pre-sixties journalism as subject to a Washington mystique that: … decreed that the people in charge in Washington knew best. They could make things happen if they wanted to. Almost all of them were acting in good faith. And they were entitled to both privacy and discretion to do what they judged necessary for the nation’s well-being.
She adds that no one believes this any longer and scarcely anyone: … admits to ever having believed it – which is a bald-faced lie for hundreds or thousands of people in Washington. So I’ll confess: I believed it. My approach to the public people I covered was that they were basically honest, competent, and usually effective. (2001: 85)
Essentially no one defends the journalism of the 1950s. Journalist Paul Duke remembered the Washington press corps after the Second World War as ‘rather sleepy’ and ‘content to report from handouts and routine news briefings’ (Sabato, 1991: 31). Reporters and politicians were frequently ‘pals’, as political scientist Larry Sabato observes (1991: 31).
What accounts for the changes in journalism in the half century from the 1950s on? Clayman and his colleagues throw their hats in with a change in the culture, norms, and values of journalism. Barnhurst and Mutz offer multiple explanations for their findings. They mention (a) the rise of investigative reporting, (b) a growing availability of social science data, the growing use of computers, in leading ‘journalists to consider the news value of what previously interested only social scientists’, (c) a growing professionalism and higher levels of education among journalists, and (d) perhaps also the growing complexity of everyday life for citizens and their greater need for assistance in seeing ‘the big picture’ (Barnhurst and Mutz, 1997).
Our new data do not disentangle these multiple possibilities. ‘All of the above’ seems the most likely answer. Still, our findings do lead us more toward general cultural explanations – growing professionalism, growing skepticism, and a growing pride in independence – rather than a specific focus on investigative and even ‘adversarial’ reporting from the Vietnam-to-Watergate era on. Although journalists have not been consistently aggressive in the past 40 years and arguably they have not been sufficiently tough, there is no sign at all of a retreat to the complaisant journalism of the 1950s and early 1960s. Every measure shows that that comfortable and cooperative journalistic world is gone.
What has replaced it is difficult to define. It is a more intellectually ambitious journalism. It is a more ‘featurized’ journalism with front-page stories of a contextual cast. It shows, as Kevin Barnhurst has observed in his impressive series of content analysis studies, ‘dramatic increases in analysis, and interpretation, if not explanation’ (2005: 253). Consistent with this trend, the Pulitzer award for ‘explanatory reporting’ was added to the Pulitzers in 1985; by the 1990s, it attracted so many entries that one of the administrators of the Pulitzers said they were getting ‘out of hand’ (Forde, 2007: 229).
Contextual reporting – a broader category that includes explanatory reporting but much more – barely turned up in our 1955 sample but by 2003 is close to half of what one finds on the front page of the three metropolitan daily newspapers we studied. The notion that the news media are dominated today by ‘he-said-she-said’ stories that write themselves is not a valid general critique of leading US newspapers, nor has it been for several decades.
Explaining this change grows more complicated still when we recognize that European journalism moved simultaneously in the same direction, even without Vietnam and Watergate. This is recounted in a careful content analysis of Swedish public broadcasting from 1925 to 2005 (Djerf-Pierre and Weibull, 2008); in a study of German campaign coverage 1949 to 2005 (reported in Salgado and Stromback, 2012); and in an account of changes toward more critical and more journalist-centered reporting in the Netherlands in the 1990s and in France from the 1960s to 1990s (see Salgado and Stromback, 2012). Whatever explanation one arrives at, it has to account for changes that affected European as well as American journalism, public broadcasting as well as commercial news output, broadcast news as well as print. This, too, moves us toward Clayman’s conclusion that a change in newsroom culture is key, more than changes fixed by particular economic, political, or technological trends or tied to specific events (Vietnam and Watergate) beyond the newsroom.
There is much more to be done to fully explain the rise of contextual journalism, but identifying the trend and acknowledging its historical weight is a vital first step. As the Nobel physicist Jacob Bronowski once wrote, ‘few things that can happen to the world are more important than the invention of a new form of prose’ (1978: 49). Scholars are well aware of the diffusion of what has been called the ‘news paradigm’ of the objective news story that originated (largely with Americans) in the late 19th and early 20th centuries (Hoyer and Pottker, 2005), but they have been much less attentive to how the dominance of the conventional news story has eroded in the past half century. Contextual journalism has emerged as a powerful and prevalent companion to conventional reporting. Its impact on how people understand their world has yet to be explored.
Footnotes
Appendix
Funding
This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.