How many Iraqi civilians have died as a result of the occupation of Iraq?
The number of Iraqi casualties has long been a topic of contention
It is a highly charged question that is now at the centre of a furious debate over the occupation's human cost, media bias and journalistic integrity.
In his recent address to Columbia University about 'Censorship by Journalism', campaigning journalist John Pilger claimed the "entire British media has omitted the true figure of Iraqi casualties, wilfully ignoring or attempting to discredit respectable studies".
Many of those opposed to the Iraq war agree, and any media organisation that now quotes certain figures is bombarded with emails.
Until recently, the most regularly used figures have come from the Iraq Body Count website, a tiny operation run by respected anti-war activist Professor John Sloboda and his colleagues.
He says he set up the site in 2003 as he was sure the occupying forces would not be keeping track of the number of Iraqi civilians killed.
He says he felt morally obliged, as a member of one of the societies responsible for the war, to set up the project.
IBC compiles a list of all violent deaths in Iraq reported in the media since the outbreak of the war.
Each report is confirmed from at least two media sources, and listed in an online database.
IBC say its database provide an undeniable and unarguable baseline figure, which those in power cannot dispute.
Because of this, those behind the site had become used to attacks from right-wing bloggers and the pro-war camp.
But for the last few months, John Sloboda and his team have found themselves in the middle of a maelstrom of criticism and vitriol.
But not from the right. Their attackers have been anti-war activists.
They stand accused of vastly underestimating the number of deaths, "handing a propaganda tool to the pro-war camp" and even "aiding and abetting war crimes".
In a recent editorial on the online 'media watchdog' site Media Lens, John Pilger described IBC's failure to respond to criticism was a "shame becoming shameful".
For a man who feels deeply opposed to the war, John Sloboda has found these attacks very painful.
"There's the emotional wearing down of having to deal with fifteen or twenty of these nasty or hostile or frightening emails every day. It makes you physically ill."
But IBC's attackers also believe they are acting out of the highest moral principles - they believe that - as they see it - 'exposing the truth' of the scale of Iraqi casualties is something that could shorten the occupation and save countless lives.
IBC say it has attempted to deflect and ignore these attacks for several months for fear of publicising a split and harming the unity of the anti-war cause, but have now decided to confront their critics by publishing a full rebuttal to the claims against them.
"Quite a lot of progressive organisations have removed, or threatened to remove their link to our work on their website, so we're losing support among the constituency that really matters to us," said Mr Sloboda.
"From the beginning, aside from the media, peace groups around the world have used our data.
"The idea that these critics will stop good people using our data is unconscionable.
This criticism has been led by Media Lens, which has run four campaigns against IBC since the beginning of the year, urging its readers to email both IBC and any media organisations who use their figures.
Its editors' central claim is that IBC underestimate the number of deaths by relying on western media reports.
Its editors say the western media routinely cover up or decline to report atrocities by coalition forces, and that the absence from the IBC database of records of large numbers of Iraqi civilians killed by coalition airstrikes in early 2005 proves that it is a partial count.
"IBC is not primarily an Iraq Body Count, it is not even an Iraq Media Body count, it is an Iraq Western Media Body Count." (Media Lens editorial)
IBC say these criticisms show a "complete ignorance of the way the media works".
"There are organisations that translate Arabic reports into English, and we see their data," said Mr Sloboda.
"One of the criticisms made is that - if only we were able to read Arabic, and could scan the Arabic media, we would find this vast array of civilian deaths that the Western media don't pick up.
"That is completely and utterly false because we have never had over the entire three years, anyone show us an Arabic source that reports deaths that we haven't already got. In three years. In thousands of incidents."
Media Lens and IBC have utterly divergent views on the accuracy of western media reports.
IBC say it depends on the 'professional rigour' of the news agencies.
IBC says it aims to be a 'public watchdog on our own governments'
"It is assumed that any agency that has attained a respected international status operates its own rigorous checks before publishing items (including, where possible, eye-witness and confidential sources)," their website states.
But Media Lens believes that the idea of an unbiased media is a myth: "'Professional' journalism has been inherently and massively biased in favour of powerful vested interests.
"It is in exactly this situation that the mainstream media become wilfully blind, wilfully naïve, and in fact function as a propaganda system for state-corporate power."
Media Lens' principal targets are the BBC, the Guardian and the Independent, organisations that they say are particularly dangerous as they claim to be independent or unbiased.
Media Lens and other critics have also attacked IBC for being run by "amateurs", unlike the professional epidemiologists who ran the sample-based Lancet study.
An editorial asks: "How many journalists are aware that IBC is not in fact run by professional epidemiologists?
"What would we say if, in discussing climate change, politicians and journalists consistently highlighted information supplied by a group deemed by professional climate scientists to be 'amateurs'?"
"Absolutely, we are amateurs," said Mr Sloboda, "rather than apparatchiks in some large organisation being paid to do this who don't care about it.
"We care about it passionately and deeply. Reading press reports, which is what this job is, requires nothing more than care and literacy.
"We believe we do it exceptionally well."
Iraq Body Count's critics who claim they are vastly underestimating the number of deaths point to a study published in the Lancet in 2004, which suggested that 98,000 civilians had died since the invasion - IBC's then count was 14,000 to 16,000.
The author of the Lancet report is respected US epidemiologist Les Roberts, renowned for his work in Rwanda, which used the same technique as his Iraq study to arrive at the widely publicised estimate of 1.7m deaths from the 1994 genocide.
The authors of the Lancet report counted the number of violent deaths from a sample of 988 Iraqi households and multiplied the results to arrive at a figure for the whole country.
Their central estimate of 98,000 has been used to suggest IBC's count is out "most likely by a factor of five or ten".
Mr Sloboda rejects this: "We've always said our work is an undercount, you can't possibly expect that a media-based analysis will get all the deaths.
"Our best estimate is that we've got about half the deaths that are out there."
The Lancet study is the highest estimate available, and other studies have suggested a lower total.
IBC's current figure for Iraqi dead is between 34,511 and 38,660
Because it dealt with a relatively small sample, the confidence interval (the range within which one can be 95% sure the real number lies) of the Lancet study was between 8,000 deaths at the bottom end, up to 194,000 at the top end.
Mr Sloboda argues this means the central estimate of 98,000 should be viewed with scepticism.
A study by the UN development agency, completed three months before the Lancet report, sampled 20,000 households and came up with a central figure for violent deaths of 24,000.
The confidence interval of this work is much smaller due to the larger sample - between 18,000 and 29,000.
IBC has now published a full examination of the Lancet study and several other studies relating to Iraqi casualties on their website.
Study the evidence yourself.