91色情片

One sunny afternoon in May four years ago, I emerged somewhat dazed from the Wellcome Trust offices in London, onto the busy Euston Road. I had popped in with colleagues for an informal chat. We walked out with an agreement that the charity would give us 拢1 million for a massive project to obtain a 7 to 8 million page collection of internal documents from British American Tobacco (BAT).

The files contained a wealth of information about the firm鈥檚 past activities and future strategies. BAT had been ordered to make them publicly available after a US court case and was storing them in a warehouse in Guildford. But the firm was making it hard for people to access them. So Anna Gilmore, Jeff Collin and I at the London School of Hygiene and Tropical Medicine, in collaboration with Stanton Glantz at the University of California, San Francisco, (UCSF) hatched a plan to get copies of every single document and put them on the internet for anyone to read.

Getting the Wellcome grant was a crucial first step. But over the next four years, we would face a succession of further hurdles. BAT made it increasingly hard for us to effectively use the Guildford files, causing long delays and obstructing our access, while carrying out covert surveillance of us at the site. But our work is now nearing completion and is already starting to yield valuable information to aid the global public health battle against the tobacco industry.

I first got involved in this project through researching the effects of globalisation on human health. While there have been increasing moves to clamp down on the tobacco industry in many western countries, cigarette sales are rising fast in the developing world. In poorer countries, where tobacco control is generally a low priority, there are billions of potential customers, few of whom are well informed about the dangers of tobacco use. These record profits keep the tobacco companies strong, enabling them to resist tighter regulation in established markets such as the UK.

Litigation against the tobacco industry has been ongoing for several decades. The Guildford files came into existence after the State of Minnesota and its health insurer sued several cigarette firms in the 1990s. The companies were legally obliged to provide the state lawyers with any relevant internal documents that could help them. They swamped the lawyers with reams of reports, letters, and memos dating from the early 1900s to the mid-1990s. As part of the eventual court settlement, the judge ruled that the thousands of files should be made publicly available for 10 years, the ones from US firms at a site in Minnesota, and the ones from British-based BAT.

The Minnesota depository, containing around 40 million pages, was managed by a firm independent of the tobacco industry, which soon put the documents on the internet for anyone to access. But the UK depository was left under BAT鈥檚 control. After almost a year鈥檚 delay, and considerable prompting, BAT finally allowed public access to the documents at a nondescript warehouse located on an industrial estate on the outskirts of Guildford.

鈥淣estled amid the golf club receipts were invoices from journalists and letters from politicians鈥�

With BAT the second largest tobacco firm worldwide, and selling over two-thirds of its cigarettes to the developing world, the public health community was eager to find out what secrets they might contain. By a lucky stroke of geography, our team at the School was well placed to take on the task of analysing the Guildford documents.

But right from the start we realised there would be problems. It was not possible just to turn up at Guildford 鈥� we had to book an appointment. We were aghast to learn that the first available weekly slot was in five months鈥� time. When the appointed week finally arrived, it was hard not to notice the CCTV cameras outside the building. More disconcertingly there were several cameras inside, and a large two-way mirror on one of the walls.

Trying to ignore this surveillance, we began working through the documents. We were not allowed into the storage area where the files themselves were kept. Instead, we were presented with a computer database compiled by BAT, which listed the titles of the almost 41,000 files containing an average of about 200 pages each. To view a file, a form had to be filled out by hand and given to a member of staff. The file would then be brought into the viewing room. We were permitted to take notes, but if we wanted a photocopy, yet another form had to be filled out. The copies would then be posted to us. This initially took several weeks, then months, and eventually over a year.

Jigsaw puzzle

We soon discovered that the indexing system. was wholly inadequate. The titles were brief and uninformative and were not organised in any systematic or logical way. Few details were given about the file contents, so there was no way to tell from the database the precise contents of each file. It would not be possible to find all documents relating to China, for example, without searching the 7 to 8 million-page collection by hand. And we were not permitted to take away a copy of the database, or to print out the results of our searches.

Despite these setbacks, as our team sat in the depository day after day, and rifled through file after file, it didn鈥檛 take long to find interesting material. Nestled amid the detritus of travel expenses and golf club receipts were invoices from consultants such as academics and journalists, correspondence with politicians, and strategies to undermine tobacco control policies in countries around the world.

We knew, however, that no one single document would reveal all. We had to follow endless paper trails to their conclusion and put individual documents in their broader context. It was like trying to put together many different jigsaw puzzles with all their pieces mixed together and without having any of the final pictures to work from.

We soon realised it would be impossible to fully exploit the archive under BAT鈥檚 restrictive terms of access. It was at this point we decided to attempt the mammoth task of ordering the entire collection file by file, and creating our own electronic archive with proper indexing and full electronic search capacity and put it on the internet. As well as securing the Wellcome Trust grant, we also secured funding from the US National Cancer Institute and Rockefeller Foundation to carry out research on the documents in relation to 20 countries and regions in the developing world.

To get the documents we had to fill out a request form by hand for each of the 41,000 files, so we drafted in further members of our research team, along with collaborators at San Francisco and the Mayo Clinic in Minneapolis, Minnesota, to help order documents.

Ordering millions of pages of documents was one thing; storing, scanning and indexing this large collection was quite another. Soon our offices were stacked high with boxes and files. At first we worried about security, but the sheer demands of finding sufficient storage space soon overtook these concerns. At one point, boxes had to be stored in the School鈥檚 ancient cellars under Store Street, where staff emptied buckets every time it rained. And it rained a lot.

We had to quickly get to grips with technical issues like which scanning formats and indexing terms to use. Invaluable advice on this came from the San Francisco team, who had already created digital archives of other tobacco industry document collections. With a job this size, we realised we would have to contract a specialist firm to scan the documents and create the indexing system. After a competitive tendering process, an American company with operations in India proved the cheapest. So the precious files ended up being shipped half-way round the world, where they were scanned and indexed, and later sent digitally to San Francisco for mounting on a website. Each time a shipment went off, I tried to suppress fears of the boat sinking.

Race against time

Our constant concern was that BAT would find out what we were doing and impose other restrictive measures to stop us. The firm was now taking over a year to provide us with the photocopied documents. With the depository due to close in 2009, the clock was ticking. BAT refused point-blank to provide some documents, saying they concerned trade secrets or attorney-client privilege. This was allowed under the terms of the original Minnesota court ruling, but we were concerned that there was no independent arbiter of whether documents withheld met this description. Indeed, the firm seemed to be using this 鈥渧eto鈥� inconsistently, as we found that some files BAT had told us were privileged, it then gave to other researchers.

We were also concerned that some files appeared to be missing. In 2000, BAT told an inquiry into the tobacco industry by British MPs that there were 40,784 files held at Guildford. But when we visited, only 40,603 files were listed on the database. What had happened to the missing 181 files? BAT has shown itself willing to destroy damaging internal documents in the past. The firm was criticised by the judge in the Minnesota court case for destroying three cases of files about the health effects of smoking.

During this period, our concerns about the project鈥檚 security began to grow. We couldn鈥檛 help but be affected by the oppressive atmosphere of the depository, and we were acutely conscious of the intrusive presence of the video cameras and the two-way mirror. While at the Guildford site, we tried to avoid discussing anything to do with our research. At the time we wondered if we were becoming overly paranoid. But last year documents disclosed during a US court case revealed that BAT staff kept records of what searches had been carried out on their database at Guildford, and ranked these searches by perceived risk to the company. The firm even kept notes about mobile phone conversations people had while they were there.

Despite our concerns, we also needed to continue fundraising. We had to explain to funders the legal parameters of the project, patiently fielding questions such as 鈥淲hy don鈥檛 you just go and ask BAT for the entire collection?鈥� One of the most difficult aspects of the project was the need to maintain strict confidentiality. Other than the most senior staff, we could not tell our colleagues what we were doing 鈥� which made explaining workload commitments tricky.

Despite our secrecy, some tobacco control advocates around the world began to hear about our project, and numerous requests for documents found their way to us: 鈥淧lease can we have everything relating to Nigeria?鈥� was a typical request. While we would have loved to have helped, it was impossible given BAT鈥檚 conditions of access. At times this gave the impression that we were somehow hoarding documents for our own purposes. Nothing was further from the truth 鈥� but how could we explain that we were engaged in secret efforts to make the entire collection available to the world? We had to remain silent despite the flak.

Smuggling

As more and more files were added to our archive, along with others, we began the task of analysing the files and important revelations are already emerging. Last month the journal Tobacco Control published five of our papers where we described BAT鈥檚 activities in Asia (vol 13, supplement 2).

The documents suggest in several countries senior BAT employees have been complicit in cigarette smuggling to boost sales and get round import bans. The documents use various euphemisms, such as 鈥渢ransit鈥� for smuggling, and 鈥済eneral trade鈥� (GT) for the contraband goods themselves. For example, in minutes from a 1991 meeting between a senior BAT executive and a Singapore wholesaler known to be involved in smuggling, the BAT staff member had asked 鈥渉ow we were progressing with efforts to increase transit鈥�. Another document described how the regional exports manager for Asia was given the specific responsibility of being 鈥渢he coordinator of GT sales worldwide鈥�, and searching 鈥渇or new GT business鈥�.

Other documents suggested that BAT had flouted tobacco advertising bans in provinces in China where the law was not yet strictly enforced. One memo, for example, stated that BAT鈥檚 goal was 鈥渢o secure quality outdoor signages for long term, in anticipation of a clamp down on major media for cigarette advertising鈥�.

Documenting these activities show how tobacco control regulations must be tightened up to prevent such circumvention. It also undermines BAT鈥檚 claims that it is a friend of the developing world by bringing investment and other economic benefits.

But these papers are just the beginning. The archive currently contains only about one million pages; as more and more files are added, it will grow at the rate of 40,000 pages per week, and should be complete by 2006. We and other groups are continuing to analyse them to see what further revelations may emerge.