BROOKE GLADSTONE: We're back with On the Media. I'm Brooke Gladstone. In the weeks leading up to war, President Bush repeatedly assured us that U.S. forces would do everything possible to limit civilian casualties, and while the military has for the most part seemed to carry through on this promise, we may never know for sure how many were killed. The Pentagon has said explicitly that it will not engage in body counts. But little by little a picture is beginning to emerge. Last week, Knight Ridder newspapers reported that according to hospital records, at least 1100 civilians and possibly twice that amount died in the battle for Baghdad alone. But that estimate was one of the very few to surface in the media so far, and some critics are asking why the story hasn't gotten more attention. Should the media be doing more? Military analyst and Los Angeles Times columnist Bill Arkin is in the process of compiling a long-term data base of civilian casualties in U.S. wars since 1991. And he says it's not the media's role to find a definitive number.

BILL ARKIN:The problem with the Knight Ridder survey of civilian casualties in Baghdad is that while this is great journalism it isn't sufficient -it isn't something that's going to stand up in history records; it would be the equivalent of the news media keeping score, literally, at sporting events and having the New York Times saying that the score of the football game was 20 to 17 while the Washington Post saying that the score of the football game was 21 to 15. The news media and journalism is the first cut of history, but it is not history.

BROOKE GLADSTONE:Since you're a bit dubious of some of the attempts by the media to provide a number based on incomplete information, what do you think the role of the media should be in reporting all of this?

BILL ARKIN:There's definitely a role for the media. In Afghanistan the, the news media did a terrific job of trying to-- give a survey of what the deaths were in the country - the Los Angeles Times - the Associated Press - the Chicago Tribune - all did great surveys of civilian deaths in Afghanistan and they expended an enormous amount of time building their own databases and using computer assisted reporting in order to do this. But the fact of the matter is that probably no one remembers what it is that they said in the end, and there's no consensus as to what the level of casualties are. The truth of the matter is that without good records at the local level and without intelligence information provided by the U.S. government, it's virtually impossible to make these kinds of calculations and estimates, and I think we're going to have that same level of difficulty in Iraq, because the United States government and the U.S. military wants to put the issue of civilian casualties behind them.

BROOKE GLADSTONE:And in the meantime, in the absence of any information from the Pentagon you have this enormous vacuum! And if the media don't fill it, and if the Pentagon doesn't fill it, you have these-- independent groups that will. If the task is left to the human rights community and fringe advocates, you know, what are we left with?

BILL ARKIN:Well I think we're left with a very bad picture, and I'm not sure that any of the non-governmental organizations or any of the news media organizations are going to resolve this issue. The Knight Ridder study is a good example. This is a great survey of the casualties and the number should be surprising to people. That study should have been front page news --not because of some war crimes issue that's lurking in the background, but because we are engaged right now in the reconstruction and reconciliation of Iraqi society! And if 1200 families lost members of their family, then those are all people that we have to deal with in the days, weeks and months ahead!

BROOKE GLADSTONE: So tell me about your project and how that fits into all of this.

BILL ARKIN:Well I think one of the things that we have tried to do is to put an academic, scientific basis for thinking about why civilians die in modern warfare, and I say modern warfare basically from Desert Storm in 1991 to the present. Because if in fact the U.S. military and the U.S. government is going to claim that it does everything that it can to minimize civilian harm, then one of the things that we need to do I studying civilian harm is trying to understand what are the major reasons that civilians die and are injured so that those practices be changed, because those practices are shown to have the greatest impact on civilian life.

BROOKE GLADSTONE:As you say, you have to take these incidents and study them individually. Do you think you're going to get enough information from the Pentagon or on the ground to be able to make reliable assessments?

BILL ARKIN:Well I think we are. I, I have been working on this now off and on for more than a decade. The Gulf War in 1991 is fairly well known. Yugoslavia is fairly well known. Afghanistan is a much more difficult case because of the lack of records at the local level, but this war probably will be one in which we'll be able to reconstruct what happened and reconstructing what happened is key in order to make these kinds of estimates.

BROOKE GLADSTONE: Exactly why do you think that number is going to be more achievable this time around.

BILL ARKIN:Well one of the reasons, Brooke, is because so many precision-guided weapons were used. With a fairly small number of tank rounds and artillery rounds and rockets being fired, we are able to look more meticulously at each of those individual weapons. We should be able to come to a fairly precise answer as to how much civilian harm and civilian death was caused.

BROOKE GLADSTONE:Numbers are rarely neutral and they're often applied in a game of blame. What sorts of applications do you see for your database?

BILL ARKIN:Well I guess our objective is to try to have a consensus document that will let people understand what the civilian cost of these various conflicts has been and give them a gauge to determine whether or not the efforts to minimize civilian harm in the future are successful or not. I don't think most people could tell you whether 1200 civilians dying in the Battle of Baghdad in accordance with the Knight Ridder study is a good number or a bad number. Did it show care on the part of the American military or was it wanton destruction? And so I would like to have the basis by which journalism and public understanding of this important issue will be improved in the future.

BROOKE GLADSTONE: Okay! Bill, thank you very much.

BILL ARKIN: Thank you.

BROOKE GLADSTONE:Bill Arkin is a columnist for the Los Angeles Times; military analyst for NBC News and senior military advisor to Human Rights Watch. He joined us from studios in his home state of Vermont. [MUSIC]