Decades ago, the folks in the computer industry developed a saying about the utility of computers: “garbage in, garbage out.” Without good data, computers have very little utility. Unfortunately, even garbage numbers can have great use as propaganda. Thus the news media feed us a constant diet of often quite meaningless statistics.
So it was with some pleasure – and some surprise — that I read an article about misleading statistics in the Washington Post. Of course, being the Washington Post, the article devoted most of its efforts to explaining the Bush Administration’s garbage statistics about Iraq. For example.
“We don’t do body counts on other people,” Rumsfeld said during the war’s ninth month, a tacit reference to the statistical excesses of Vietnam. Yet that rule has been bent repeatedly, with scant explanation of how figures are compiled.
On April 6, 2003, the Pentagon listed 61 U.S. soldiers and Marines killed or missing in action, while officers in the midst of battle estimated that “2,000 to 3,000″ Iraqi combatants had been killed during a single tank incursion into central Baghdad. A year and a half later, as the U.S. death toll topped 1,000, Rumsfeld observed that in August 2004 alone, U.S. and coalition forces “probably” killed between 1,500 and 2,500 terrorists and criminals. In January 2005, Army Gen. George W. Casey Jr., then the top U.S. military officer in Iraq, estimated that U.S. and Iraqi forces had killed or captured 15,000 enemy fighters in 2004 — three times as many as an estimate by Gen. John P. Abizaid, then-Centcom commander, of the total size of the insurgency about one year earlier.
The number of enemy losses, Pentagon spokesman Bryan Whitman told The Washington Post about 2 1/2 years into the war, is “a metric that can help convey magnitude and context” after a battle. Military officials said the release of such numbers helps bolster the morale of U.S. forces. But there is no way for outsiders to verify them.
However, let’s give credit where some is due. In this case the Post exceeded in some small small degree its usual halfhearted effort to appear evenhanded. This article actually slighted that sacred number of 600,000 dead Iraqi civilians.
Civilian casualties have proved even harder to pin down. Before the war, the United Nations predicted that they could reach 500,000; once it started, critics claimed thousands of civilian deaths. But in his May 1, 2003, “Mission Accomplished” speech, Bush said the use of precision weapons had largely diminished noncombatant deaths.
On Dec. 12, 2005, Bush offered his first and only number of Iraqi casualties: “30,000, more or less,” without distinguishing between enemies and noncombatants. The independent, London-based Iraq Body Count offered a similar figure for civilians only — 34,516 to 38,661 Iraqis dead by early 2006.
The group’s most recent tally, drawn from global media accounts, estimated civilian deaths between 59,287 and 65,121, as of yesterday. The British medical journal Lancet estimated 100,000 civilian casualties in the 18 months after the invasion, and in October it raised its total to 600,000. On the low end, the independent, U.S.-based Iraq Coalition Casualty Count placed the one-year total from March 2006 to last weekend, including Iraqi security forces, at 21,186.
The United Nations, using reports from Iraqi morgues, hospitals and local authorities, placed the 2006 figure at 34,000 — three times the official Iraqi government count. Iraqi Prime Minister Nouri al-Maliki’s office ordered the country’s health ministry to stop providing figures to the United Nations.
The Defense Department does not release detailed tallies of Iraqi casualties, but in an Iraq security assessment published last week it said the number decreased in January “but remained troublingly high.” Noting that only incidents “reported to or observed by Coalition forces” were included, it said that the U.N. estimate of more than 6,000 civilians killed or wounded in December was “about twice as many casualties as were recorded by Coalition forces.”
So what should we make of the numbers? I suggest skepticism. Even when working with a peaceful, cooperative population, statistical data is notoriously hard to gather. Given that most people in Iraq have greater concerns than gathering accurate data for statistical analysis, we have garbage going in. Inevitably, we will have garbage going out.
Sometimes we have to go with our gut and past experience. Since few of us are in Iraq and even fewer of us have past experience in nation building, that is why we all need to listen to people on the ground and to read our history books.
