The newsworthiness of death

Why did the Mercury News and Associated Press give Israeli deaths greater prominence than Palestinian in the Middle East conflict?

Analysis by John McManus
Posted December 18, 2003                              How the study was done

On April 28 last year, the San Jose Mercury News reported the murder of four Israeli settlers. The graphic description could only provoke outrage:

Palestinian gunmen dressed as Israeli soldiers slipped into a quiet Israeli settlement near Hebron on Saturday morning and, moving from door to door, sprayed bullets through windows and walls and killed four people. The gunmen entered the home of 5-year-old Danielle Shefi, who lived with her parents and three siblings. After they left, the little girl's room -- decorated with a Mickey Mouse doll, bed sheet and poster -- was scarred with bullet holes and stained with blood.

The 1,151-word story ran on the front page.

Four months later, on Aug. 29, the Mercury News reported on four Palestinian deaths:

A Palestinian woman, her two sons and a cousin were killed early today when Israeli tank shells exploded in a Bedouin encampment near an Israeli settlement in Gaza City, residents and doctors said. Four others were wounded, including the woman's 4-year-old son, said doctors at Shifa Hospital in Gaza City.

The 187-word story ran on page 11A.

Beyond the difference in size and prominence, note the attention to humanizing detail in the first report, and the lack of attribution -- no "said's," just facts. The second report of the killing of four other victims of the Israeli-Palestinian conflict is carefully attributed to Palestinian "residents and doctors."

Differences in two isolated stories of death could have any number of journalistically defensible explanations. But a study of six months of the Mercury News' coverage during the height of this conflict from April 1 to Sept. 30, 2002, shows that an Israeli death was 11 times more likely to make a front-page headline in San Jose than a Palestinian fatality.

In fairness to the Mercury News, if you looked at the smaller text on the front page and particularly if you followed the stories inside the paper and read the adjoining "sidebar" articles, the numbers came closer to balance. And there were obvious instances on inside pages where editors struggled to match accounts in which each side described its victimization. Although outside our analysis, editorials also appeared to treat both Israelis and Palestinians with sympathy and respect. We also noted that photos of grief were often paired.

Reporters covering the struggle neither write the headlines nor decide which story goes on 1A for all to see, and which goes inside where readership wanes. Editors prioritize stories and boil down their import to a sentence or two.

To the editors at the Mercury News, the story was much more about Israeli deaths than Palestinian, even though during those six months 499 Palestinians and 192 Israelis died in the conflict. Over the period, Mercury News page-one headlines reported 147 Israeli fatalities and 35 Palestinian, 77% and 7% respectively.

Wording and placement

The priority given to Israelis suffering in front-page headlines was reinforced throughout the stories by the order in which casualties were described, and first-person accounts of Israeli deaths contrasted with second-hand and approximate estimates of Palestinian fatalities. We also noticed that Palestinian forces were consistently labeled as "gunmen" and "militants" -- terms with negative connotations in our culture -- but rarely as "fighters" and never as "resistance forces."

Because some Mercury News articles were first reported by the Associated Press, and the AP serves so many American newspapers, we also analyzed every story the AP moved during the study period using the Lexis-Nexis database.

The analysis of Associated Press coverage could not distinguish between front page and inside stories, since those choices are made by editors at each paper using the service. In its headlines, Israeli deaths were twice as likely to be mentioned as Palestinian. When we included the top five paragraphs of each article -- the section most likely to describe fatalities -- the ratio narrowed. Israeli deaths were 37% more likely to be reported than Palestinian.

Both analyses relied on counts of actual Israeli and Palestinians killings supplied by the Israeli human rights organization B'Tselem. It was established in 1989 by a group of prominent Israeli academics, attorneys, journalists and members of Israel's parliament, the Knesset. As a check, the Associated Press provided Grade the News its internal and independent count of fatalities. According to corporate spokesman Jack Stokes, the AP checks every death related to the conflict by interviewing relatives, witnesses and doctors, and visiting hospitals. The AP and B'Tselem tallies are nearly identical over each of the six months. The AP counted 196 Israeli and 544 Palestinian deaths.

First impressions matter

Front-page headlines exert a disproportionate influence on public attitudes, according to William Woo, a journalism professor at Stanford University, veteran newspaper editor and advisor to Grade the News. "If all I'm seeing are Israelis being killed by Palestinians," he said, "I'd be likely to conclude that the Israelis are the predominant victims of violence there."

"The headlines do have a lot of weight," said Rashid Khalidi, professor and director of the Middle East Institute at Columbia University. "A lot of people, all they'll ever see is that headline. I think it has a big effect on public opinion.

"When most of the victims were Palestinians, [American] press coverage was much lighter," Prof. Khalidi said. "When Israelis started being killed in the spring of 2001, you then got a very different reaction."

"If it has been documented that Israeli deaths were considered more newsworthy than Palestinian," commented Stanford Communication Prof. Shanto Iyengar, "that's prima facie evidence of bias."

Not so, protested Daniel Sneider, who was foreign and national desk editor at the Mercury News during the study period. He refused to say why on the record. But Mr. Sneider called a similar study to ours, conducted by an organization called If Americans Knew, "fundamentally flawed." Grade the News replicated and expanded the study conducted by the Berkeley-based media monitor. Because we included Palestinian deaths implied by the term "suicide" in our totals, Grade the News showed slightly less imbalance than If Americans Knew. Otherwise, our counts matched theirs.

Two other top editors we contacted at the Mercury News declined comment, deferring to Mr. Sneider, who now writes a foreign affairs column for the newspaper.

David Yarnold, who was executive editor at the time, defended the Mercury News' coverage in an April 7, 2002, article in the paper's Sunday Perspective section. "In 25 years in journalism," Mr. Yarnold wrote, "I've learned that no story is more volatile than the Mideast and never more than now.

"First and foremost," he explained, "we try to cover events fairly and without bias." He noted that editors read accounts of the conflict from a variety of newspapers and wire services, not just reporting from corporate parent Knight Ridder's Jerusalem bureau.

"Second, we strive for balanced presentation. There are two sides in this long conflict, two peoples, and the essence of this conflict is that they're inseparable."

An Associated Press reporter stationed in Israel expressed surprise at the differential in reporting Israeli and Palestinian deaths. The reporter would only speak on the condition of anonymity, citing a corporate rule forbidding interviews. "We are always examining ourselves. We don't necessarily report every death." But the ratio of those reported to those killed should be in balance between the two groups, the reporter said. "It's our basic interest to be fair and balanced."

If both the Mercury News and Associated Press seek fair and balanced reports, why do Israeli deaths merit greater attention than Palestinian?

Myriad possible sources of bias

Academics, partisans of both Israel and Palestine, and journalists -- some of whom have reported the conflict -- offer several explanations. Among the most prominent are:

• Israeli victims are predominantly innocent -- civilians heading to work or sitting at a cafe -- while Palestinian victims are predominantly combatants. The killing of innocents is more unusual and more deserving of public attention than the killing of combatants.

• Access. Most American journalists are stationed on the Israeli side of the "green line" and can get to the scene of an attack on Israeli civilians more readily than an Israeli Army assault on a Palestinian city. The army often blocks journalists from covering its attacks. Further, the Israeli government is more organized and better at putting its version of events before the press than the Palestinian Authority, which was cut off in Ramallah during most of the study period. Language barriers on the Palestinian side also limit access. American reporters generally rely on translators to speak with Palestinians, but many Israelis speak English.

• Cultural and political affinity with Israel. American editors -- and Americans in general -- see the Israelis as more like themselves than the Palestinians. American foreign policy, especially under the current administration, supports Israel over the Palestinians financially, militarily and politically.

• Pressure from the local community. While editors hear from both sides, the partisans of Israel are more organized and influential than supporters of Palestine.

• American journalism is event-driven rather than issue-driven. The number of Israeli victims in a single incident generally exceeds the number of Palestinians killed at one place and time. Two or three people killed overseas may not merit the front page, but a dozen do.

Innocent victims' deaths deserve the front page

Writing in the May/June issue of Columbia Journalism Review, Ira Stoll argued the first point: "The moral issues are so clear-cut -- innocent Israeli bus-riders are being deliberately targeted in cold blood by terrorists funded and encouraged by brutal tyrants -- that it makes balance difficult."

In this view, most Israelis killed are victims of Palestinian terror attacks. When the Israeli Army kills Palestinians, it does so in national self-defense. If innocent Palestinians are killed along with combatants, their deaths are accidental. News media act appropriately when they distinguish between aggressors and victims.

Prof. Khalidi responded, "All the independent observers have concluded that the very large majority of those killed on both sides are innocent victims."

Who is innocent depends on point of view

Jason Keyser, an Associated Press correspondent in Jerusalem wrote, "the death toll is itself part of the Israeli-Palestinian struggle, with each side contesting the other's version to portray itself to world public opinion as the greater victim."

In an article published just after the study period of our analyses, Keyser noted that "Israel's Foreign Ministry maintains no more than 45 percent of the Palestinian casualties of the conflict were 'noncombatants,' people who were neither involved in hostilities against Israel nor members of Palestinian armed groups. But the Palestine Monitor, a think tank that tracks the violence, says 85 percent were 'civilians' -- and appears to include in that category all those who were not members of the Palestinian security forces, even if they belong to various armed groups."

Even if the Israeli Foreign Ministry's numbers are accepted, however, up to 45% of Palestinian victims may be non-combatants who presumably qualify as innocent. So innocence would not go very far in explaining an 11-to-1 priority for Israeli victims in Mercury News front-page headlines.

Access

Mercury News reporter Elise Ackerman covered the conflict in the spring of 2002. "During March and April it was really difficult to operate as a reporter," she said. "The Israelis closed areas [in the West Bank] off in an unpredictable fashion. You had to find a way around their roadblocks. Local fixers know ways around, but it's hard to take the time when you're on deadline, and it's dangerous."

Palestinian sources, she explained, would make themselves available, but "they didn't necessarily have the information because their government wasn't functioning." At the time Palestinian leader Yasser Arafat was isolated by Israeli troops at his compound in Ramallah.

Daniel Rubin, a Philadelphia Inquirer reporter, covered the conflict at the same time for the Jerusalem bureau of Knight Ridder, the corporate parent of both the Inquirer and the Mercury News.

"How good were the Israelis as manipulators? They're really good," he said. "They were professionals. But the Palestinians are really good too." Both sides, he said, were eager to show reporters what the other had done to them.

When the Israeli Army blocked press access, Mr. Rubin said, he sent his translator in. "We worked really hard to make sure everyone who was suffering was represented."

Reporters had an easier time reporting the Israeli side, according to Mr. Rubin and other journalists.

In an article titled "In Israel, Press Kits Roll Out With Tanks," New York Times reporter John Kifner described an information offensive parallel to the Israeli offensive launched in late March 2002. "Even as the tanks rumbled into Ramallah this morning, Gideon Meir, Israel's deputy foreign minister for public affairs, was setting up a huge government information office in Jerusalem's Convention Center to provide daily briefings," he wrote.

"Workers bustled about today, setting up tables, computers and telephone lines. On a long row of tables, there were handouts ranging from maps and pamphlets with basic facts on population to piles of CD-ROMs. The discs included 'Surviving Terrorists' ... and 'Seeds of Hatred.'"

The language barrier

Ms. Ackerman describes language as a second barrier. "It's really hard to find good Arabic translators," she said. "It creates a problem in reporting if you can't call someone and talk to them in English. The AP had Arabic-speaking reporters but we didn't" in Knight Ridder's Jerusalem bureau. The bureau had to rely on temporary employees who were not trained as translators and not always available, she said.

"People tried to be balanced," Ms. Ackerman explained. "But they are not getting a sense of the real texture of the story. There's less 'hanging out' kind of reporting about the Palestinians. Israel’s society is a lot more accessible."

Reporters can write more descriptively and authoritatively when they witness violence than they can when interested parties, or indigenous employees, tell them about it second-hand. That would explain why first-day reports of Palestinian deaths were usually estimated or described with terms like "at least" and "approximately" while Israeli death counts were precise.

Differential access may also explain why death tolls were distorted, under-reporting Palestinian deaths. But reporters were free to visit Palestinian areas after the Israeli Army left and provide fuller accounts. These accounts, however, almost never made the front page of the Mercury News during the study period.

As the newspaper chain with the second largest circulation in the United States, Knight Ridder can afford trained translators and it has the capacity to hire Arabic-speaking journalists.

Cultural affinity with Israel

Marda Dunsky, an assistant professor at Northwestern University's Medill School of Journalism, is researching a book about how American news media report the Israeli-Palestinian conflict. In her view, objective standards of news judgment sometimes are trumped by two factors. The first is that Americans tend to identify with Israel and Israelis more than with Arab nations and Palestinians.

Israel is a frequent travel destination for Americans, especially devout Jews and Christians who consider it the "Holy Land." Israel also comes closest to democratic ideals among Middle Eastern nations. Prof. Dunsky points out the long-standing political relationship between Israel and the United States. Because American foreign policy treats Israel as a special ally, she explained, Americans tend to as well.

Cultural ties between Americans and Israelis can also affect the language choices reporters make, according to Prof. Dunsky. The name of the Palestinian group HAMAS is an acronym, she explained. The M stands for "mukawama" which is translated as "resistance." Yet Palestinian fighters are almost never referred to as "the resistance."

"I didn't use the word resistance," said Ms. Ackerman of the Mercury News. "I couldn't believe that I hadn't used the term resistance in my stories. There's been a resistance throughout the Arab world. It's a resistance to colonizers.

"I think that whole idea is missing from the coverage of the Palestinian conflict because there is a resistance of Americans and Israelis to see the Israelis as colonizers and that creates a problem with the use of the term."

“I do know in hindsight that the Palestinians do see this as resistance. The translator may not have provided the nuances. In hindsight I think I did really get lost."

Pressure from the local community

Prof. Dunsky said the second reason objectivity standards haven't always applied to the conflict is pressure from local groups. Pressure comes from both sides, she said, but it "tends to come in a much more organized, vociferous and frequent manner from the Jewish-American community than its Arab/Muslim counterpart."

As evidence, she cited an interview with an editor at a major newspaper after the paper had come under fire from the pro-Israel faction. "The desk was given very explicit instructions about care to be taken with editing. The editor recalls joking with colleagues, 'We can say, 'Palestinian walks in front of Israeli bullet,' but not, 'Israeli bullet kills Palestinian.'"

Prof. Khalidi said American newspapers are targeted by pro-Israel groups whose politics are far more partisan than most American Jews' if they report fairly on the conflict. "These are organized campaigns, backed by millions of dollars. It's not just that a lot of people feel strongly attached to Israel, but there's also manipulation and hysteria and cynicism."

As evidence of the level of organization and coordination of the pro-Israel message, Mr. Khalidi provided a copy of renowned Republican pollster and message-crafter Frank Luntz's "Ten Commandments of Effective Communication." The 36-page manual is "based on a comprehensive four-month nationwide research effort of polling, instant response dial sessions and focus groups."

Among the "commandments" are injunctions to convince Americans that violent Palestinian opposition to Israel is equivalent to al Qaeda's attack on the Pentagon and World Trade Center on Sept. 11, 2001. "Explain why a threat to Israel is a threat to America," reads commandment four.

Commandment two urges readers to conclude every discussion of the issue with a specific reference to 9/11 and provides this script: "On that fateful day, Israelis shed tears of pain for the Americans who were killed. But on that day, the Palestinians danced in the streets in celebration. Yes, there is a difference between the two peoples."

No editor we approached at the Mercury News would speak about pressure on the newsroom or its effects on coverage. But Ms. Ackerman said, "Dan Sneider would add things to my story to be sure they were balanced. They tended to be balancing things to be sure the Palestinian side got in. He [did] bend over backwards to be fair as an editor."

Events with multiple deaths merit the front page

To investigate whether the Mercury News was following a newsworthiness standard of numbers killed -- regardless of nationality -- in a particular incident, we looked at the average number of deaths per news report about fatal Israeli-Palestinian violence from the broadest sample we could find -- the output of the Associated Press. We identified 77 such stories in the six-month study period.

The average number of Israeli deaths per incident reported was 4.6; The average number of Palestinian deaths was 2.9. At least in the eyes of the AP, more Israelis died in incidents considered newsworthy than Palestinians. But the ratio -- about one and a half to one -- again falls short of the difference between Israeli and Palestinian fatalities on the Mercury News' front page, both in headlines and text.

"It is true in a certain way," Mr. Khalidi said. "Ten people being killed on a single day has a certain political impact. But it means that all that gets to the top is a few high-profile events. Yes, the 10 people killed is very important, but the fact that it came after 18 Palestinians were killed in a sweep ... would give context."

In April 2002, an Israeli offensive was still underway in the West Bank. According to B'Tselem, 247 Palestinians were killed by Israeli troops, including 26 under the age of 18. Not one was mentioned in a Mercury News front-page headline in April, but three headlines described Israeli fatalities.

Even Israeli combatants rated a front-page headline. On April 10, the paper reported, "Deadly ambush on army kills 13 Israeli soldiers." Yet in the body text, the paper reported, "Palestinian witnesses estimated as many as 150 Palestinians have been killed this week in Jenin in the fiercest fighting yet in Israel's 12-day-old offensive." If sheer numbers of fatalities drove the Mercury News' headlines, 150 would trump 13.

Given the professionalism of the journalists both at the Mercury News and the Associated Press and their surprise when confronted with an overview of their reporting, it's unlikely that the priority given Israeli suffering over Palestinian in the six-month analysis period was deliberate or a conscious acquiescence to local pressure. But editors are as susceptible to bias as anyone, particularly when it's embedded in the culture and taken for granted.

Rajeev Poduval, a volunteer with Grade the News who spent time reporting in the Middle East, helped research this story.

How the study was done

The idea for analyzing the San Jose Mercury News' coverage of the Israeli-Palestinian conflict came from a study published by a Berkeley-based group, If Americans Knew. That analysis of front-page headlines, between April and October, was startling, claiming that 73% of Israeli fatalities were noted, compared with just 5% of Palestinian deaths.

Frankly, the disparity seemed unbelievable, so we decided to replicate the study ourselves and expand it to look beyond headlines, to the text of the stories on the front page. Less formally, we looked at inside stories as well.

We also included the Associated Press during the period to provide a parallax view.

We counted only unique mentions of deaths on either side, and only those the paper linked to the fighting. So if a headline summarized deaths over a week of violence, we only counted those previously unreported. We did not count foreign nationals killed, so our percentages would align with those of B'Tselem, the Israeli human rights group whose casualty tallies we relied upon.

Because some headlines mentioned fighting, but not numbers killed, we analyzed the first five paragraphs of each story beginning on the front page. In almost every case this included all of the text of the story on the front page, before it continued on an inside page.

The Mercury News' headlines mentioned 147 of the 192 Israeli deaths recorded by B'Tselem during these months, or 77% of the total. Just 35 of the 499 Palestinians slain, or 7%, were mentioned in headlines, even when we included those implied by the word "suicide" but excluded from the number of dead reported in the headline.

Front-page text in the Mercury News expanded the percent of Israeli deaths mentioned slightly to 88%, and Palestinians more substantially, to 54%. But a single story, of the 140 analyzed, accounted for more than half the mentions of Palestinian deaths. Absent that article, fewer than one in four Palestinian deaths were mentioned in front-page text.

Israeli deaths are reported high in the story

Since readership declines with the length of stories, reporters put the most valued facts high in the story; less important ones come later. It's called "inverted pyramid" style. Israeli deaths almost always were reported before Palestinian deaths, even when the number of Palestinians killed was far greater.

The Mercury News' front-page story on April 10 illustrates the pattern. The main headline read, "Deadly ambush on Army kills 13 Israeli soldiers." The story's lead paragraph elaborated on the ambush. The second paragraph described a Palestinian suicide bombing killing 10 Israelis on a bus. The third paragraph returned to the ambushed soldiers. The first mention of any Palestinian deaths came in the fourth paragraph, an estimated 150 killed in the invasion of Jenin. Their deaths were not mentioned again until late in the story, as the focus returned to the ambush and the bus bombing.

On April 3, the deaths of 13 Palestinians weren’t reported until the ninth paragraph of the story, well after it jumped from the front page. On April 8, the killing of 14 Palestinians wasn’t mentioned until the ninth paragraph. On May 5, the Associated Press reported that Israeli soldiers killed a Palestinian woman, her two children and a young boy who were picking grape leaves in their yard. The Mercury News mentioned the incident on May 7 on page 12A in a summary box.

Far fewer Israeli deaths triggered front-page mention. On May 20, a suicide bombing claiming three Israeli lives made page one. Three days later, a 1A article described a bombing that killed two Israelis. On July 13, the deaths of three Israelis in a suicide attack were reported on the front page.

The picture is not black and white, however. An Israeli tank firing on a crowded market rated a rare front-page headline in which Palestinian deaths were mentioned. And on July 23, when an Israeli plane fired a missile at an apartment building to kill a Hamas leader in Gaza, the deaths of Palestinian children were reported on the front page.

Other studies

Other studies of the Mercury News have come to varying conclusions. If Americans Knew followed up on its first study of the newspaper for another six months and found results similar to its first study -- a strong bias in front-page headlines toward the Israelis. The group has also rated the San Francisco Chronicle's coverage of the conflict.

In September 2002 another group -- the Jewish Community Relations Council of San Francisco, the Peninsula, Marin, Sonoma, Alameda and Contra Costa Counties -- issued a study of the Mercury News' coverage of the Middle East charging bias in the opposite direction.

Studying the period from June 19 to July 24, 2002, the Council concluded, among other measures, that the Mercury News published seven times more photos of Palestinian victims than Palestinian combatants and more pictures of Palestinian victims than Israeli victims.

-- John McManus