As the scale of the contamination became known, some refused to believe it could be true. Others were shocked — and angry. When plaintiffs' lawyers from as far away as Pennsylvania began recruiting clients in the late 1980s, more than 5,000 signed on to sue the government contractors that ran the plant, alleging that the radiation releases caused cancer and other illnesses.

This is not a story, however, about a group of victims who triumphed against the government. After 21 years of litigation, only a few downwinders have received compensation. The case remains bogged down in motions and competing expert studies, and prospects for the remaining 1,500 plaintiffs to receive any money in their lifetimes are dicey.

Part of the problem is the case's unwieldy size and high stakes. But blame for the costs and delays also falls on the lawyers and judges themselves, who over the years — a review of hundreds of pages of court records shows — have made errors in strategy and judgment, staking out positions that leave little room for compromise.

On one side are the defense lawyers, who have billed the government more than $50 million but say most of the plaintiffs don't deserve anything because they can't prove their illnesses were caused by Hanford radiation. "We're not going to throw taxpayer money at baseless claims," said lead counsel Kevin Van Wart, a partner at Kirkland & Ellis in Chicago.

The plaintiffs' lawyers handling the case on contingency have spent nearly $10 million on litigation expenses — never mind thousands of hours of as-yet unpaid work. At this point, they can't afford anything less than a massive payout from the government if they hope to break even.

Settlement offers by the U.S. Depart­ment of Energy to date — $10,000 for a case of thyroid cancer, for example — are insulting, they say.

As for the judges, the first was forced to recuse himself midway through the case due to a conflict of interest. The second has only recently taken steps to speed up the proceedings to allow some plaintiffs a day in court.

In the meantime, the downwinders are getting older and sicker — and dying empty-handed. "I've resigned myself to the fact that this is never going to be resolved in my lifetime. It just grinds on year after year after year," said Means, who is 71. "I have three children, and I hope it results in something for them."

Diagnosed with chronic fatigue syndrome and thyroid disease, she has felt sick — flulike, achy and tired — for most of her life. The only consolation to learning about what happened at the plant is that she felt her condition was finally explained. "I have classic radiation symptoms," she said. "It was like a light went off in my head. 'Is that what's been going on?' "

Now she sits alone on a wooden bench outside a ninth-floor federal courtroom in Spokane. She's the first to arrive for a status conference before U.S. District Senior Judge William Fremming Nielsen. She and about a dozen other plaintiffs have come to observe the proceedings (routine as they may be), a reminder to the lawyers and the judge that they are here and they are waiting. "They radiated the whole place," she said, her pale blue eyes welling with tears. "To think that your country would do that to a person and deny any wrongdoing."

Hanford is about 170 miles southwest of Spokane, in a remote desert scrub landscape of hills and tumbleweed. The nearest town is Richland, where local businesses have names like Atomic Auto Body and Atomic Bowl, and the high school mascot is the Bombers. The Army Corps of Engineers' Manhattan Project, racing to develop a nuclear weapon during World War II, picked the Hanford site in 1943 in large part for its proximity to the Columbia River, since water was necessary to cool the reactor cores.

The government knew it didn't have the expertise to run the plant itself, so it recruited E.I. du Pont de Nemours and Co. The company initially refused. Only after the government insisted that the project was of utmost importance to the war effort did DuPont agree out of "patriotic considerations." The company accepted one dollar as payment for its services, according to court papers.

In September 1944, Hanford began making plutonium, and continued to do so for the next 40 years.

DuPont soon transferred its duties to General Electric Co., which ran Hanford through the end of the Cold War and also agreed to earn no profit for its work.

The contractors may not have made money off Hanford, but they weren't going to lose any, either. As part of the agreements to run the plant, the government indemnified them from all liability. Contractor immunity was later incorporated into the federal law that covers nuclear accidents, the Price-Anderson Act. That's why, although GE and DuPont, not the United States, are named as defendants in the suit, the government is on the hook for damages. It's also why the companies got to hire their own lawyers, again paid for by the government. Kirkland & Ellis, originally retained by DuPont, is the only one still remaining. Neither DuPont nor GE is actively involved in the day-to-day litigation, which is managed by the general counsel's office in the Department of Energy.

Kirkland has been paid $21.4 million in fees and $2.7 million in expenses by the government for its work from 1990 until the end of 2010, according to Van Wart.

To lead plaintiffs' counsel Louise Roselle, a partner at Cincinnati's Waite, Schneider, Bayless & Chesley, this arrangement is one reason why the case hasn't settled. "When the U.S. government is paying the bills, you lose a huge check and balance," she said. "A company looks at what the litigation is costing, and makes a cost/benefit decision. The government doesn't seem to care."

There's also the attitude — what Roselle and other plaintiffs' lawyers have called a "scorched earth defense." Kirkland lawyers have a reputation for being tenacious and aggressive litigators, and Van Wart is no exception. It's an approach that plaintiffs complain is at odds with the Price-Anderson Act's mandate for the government to "provide for full and prompt compensation of all valid claims" from nuclear incidents.

The environmental contamination at Hanford began almost immediately after the plant opened in 1944. Until 1948, there were no filters on the smokestacks. Radioactive gases and particles flew straight up the stacks and into the air.

Many of the substances were released in small quantities or became nonradioactive very quickly, but not iodine-131, which is produced by the fission of uranium atoms.

In 1945 alone, the plant released an estimated 555,000 curies of radioactive iodine, compared with 15 curies during the Three Mile Island accident in Pennsylvania in 1979. (A curie is the number of atoms that decay each second. One curie is 37 billion atoms.)

People as far away as Idaho and Oregon were exposed to iodine-131 by eating contaminated fruit and vegetables, breathing contaminated air and drinking contaminated milk. One of them was Jay Mullen, a plaintiff in the case. From 1944 to 1945, when he was 4 years old, he lived with his family at the Farragut Naval Training Station in Idaho, 225 miles northeast of Hanford. Farragut lay on the outer reaches of the fallout area, but Mullen drank milk — big glasses of it at daycare on the base, where you couldn't leave the table unless you'd finished every drop.

Cows can be highly effective transmitters of radioactive iodine. When radioactive material falls on pasture grass eaten by cows, it passes into the cow's milk. Once in the human body, iodine-131 concentrates in the thyroid gland, where it can eventually cause thyroid disease and cancer. Because children's thyroids are smaller, they absorb proportionately more iodine-131 and the harm is much greater.

Fourteen years later, Mullen, by then a college athlete at the University of Oregon, woke up one morning and couldn't move. "They concluded I had thyroid-induced paralysis," he said. "I was told I was a medical freak. They couldn't explain it."

Today, there's no doubt in his mind that his condition was caused by radiation from the plant, ingested via local milk. His thyroid was removed, and he recovered — but he still has a scar at the base of his throat that he calls his "Hanford necklace," and he's had to take thyroid medication every day for the past 50 years.

The most egregious release of radiation, the "green run," came in December 1949. A few months earlier, the Soviet Union detonated its first atomic bomb, and the Air Force wanted to find out how an airborne release of radioactive materials would spread — data that could be used to monitor Russian nuclear weapons programs, according to a 1992 government-funded report.

Hanford scientists intentionally released about 7,800 curies of iodine-131 over 12 hours, according to the report. No one in the community was informed, though plant operators had some inkling the substance was harmful, since the scientists who designed Hanford had set tolerance doses for human exposure.

Contamination from the plant affected the Columbia River as well. River water that cooled the reactors was passed through the plant and discharged downstream. In one instance, a dike broke and 28 pounds of uranium spilled into the river.

Over the years, as technology developed and safeguards were put into place, the amount of radiation released by the plant dwindled to almost nothing. In total, from 1944 to 1972, the facility released about 739,000 curies of iodine-131, the vast majority of it in the first five years, though plaintiffs contend that the amount is likely even higher.The Columbia River pollution peaked in the late 1950s to mid-1960s.

The amount is far less than Chernobyl, which released more than 10 million curies of iodine-131, or the Fukushima plant in Japan, which has released an estimated 2.4 million. The difference, though, is that the people in the Soviet Union (at least to some degree) and Japan were aware of what was happening. If the danger is known, the absorption of radioactive iodine can be blocked by taking potassium iodide — but the people near Hanford knew nothing for 40 years.

The disclosure began in 1986, when Hanford was one of three final candidates for a permanent nuclear waste dump (Yucca Mountain in Nevada was chosen instead). At a 1985 community forum to discuss concerns about potential health hazards, Michael Lawrence, who was the Hanford plant manager for the Department of Energy, promised to release records about past emissions in what seems to have been a misguided attempt at reassurance. On Feb. 27, 1986, Energy officials released 19,000 pages of documents, including partial data about the radioactive releases.

Press reports and public outcry followed — and so did lawsuits.

In the early 1990s, radiation cases were proliferating. The Department of Energy had just settled a class action filed over contamination from its Fernald uranium processing plant in Ohio for $78 million. Roselle's firm, Waite Schneider, was lead counsel. In a later Fernald case, Van Wart was among the opposing counsel.

Another Hanford plaintiffs' firm, Phil­adelphia's Berger & Montague, handled litigation stemming from the Three Mile Island accident. The firm went on to work with Waite Schneider in suing the government over pollution from the Rocky Flats nuclear weapons plant in Colorado. Again, Van Wart was on the other side (Years later, he thanked both firms in court for creating "a new practice area for me and my firm. I appreciate it.").

As a series of personal injury claims rather than a class action by property owners, Hanford was bound to be more complicated than the other cases. It was also potentially more lucrative — according to the defense, the plaintiffs wanted $2 billion to settle the matter, though plaintiffs insist they never asked for more than $500 million.

The cases, In re Hanford Nuclear Reser­vation Litigation, were consolidated in the U.S. District Court for the Eastern District of Washington in 1991. The joint complaint alleged violations of the Price-Anderson Act, charging that the defendants acted intentionally or negligently, and asked for compensatory and punitive damages for physical, emotional and economic harm. The matter was assigned to Judge Alan McDonald, who wrote that "the potential enormity of this litigation, as well as the dollar amount of any recovery, is almost staggering."

To attract would-be clients, some plaintiffs' lawyers went town to town holding local meetings, put up kiosks in shopping malls and advertised — but didn't always do much in the way of vetting. Some who signed on, it turns out, didn't live in the area when the radiation releases occurred.

Still others complained of ailments like bad teeth and gray hair that weren't caused by radiation. Another plaintiff, in an interrogatory, described her symptoms as follows: "As I would face the street light on the corner where I was standing, it would go out. I attribute that to my being radioactive."

"It was like a scene from the Wild West," Van Wart said. "Their only focus was on enrolling plaintiffs." He attributes some of the case's subsequent delays and costs to the bloated roster of clients. "By their indiscriminate claims-solicitation practice, the plaintiffs' lawyers dug themselves into a huge financial hole," he said.

Roselle dismisses this as little more than "defense rhetoric." At the time, she said, it wasn't clear exactly who had been exposed to what. "People were understandably concerned, and would go to a lawyer to get representation."

Still, plaintiffs' lawyer Richard Eymann, a name partner at Eymann, Allison, Hunter, Jones in Spokane who became involved in the case after the rush to sign clients, does concede that "some of the claims were not legitimate."

Also, many of the original claims were from people who were not sick, but who wanted medical monitoring. They were dismissed early in the proceedings.

The plaintiffs' lawyers — there are nearly three dozen of them listed on court papers — initially asked that the case be certified as a class action, a request that they eventually withdrew, recognizing the injuries were too individual to be lumped together.

It would, however, have simplified the lawyering immensely. If Hanford had been a class action, the judge would have appointed lead class counsel, selecting a lawyer with the expertise and resources to call the shots.

Instead, the plaintiffs all picked their own lawyers. The one most of them preferred, who empathized and listened and kept them informed, was Nancy Malee Oreskovich.

Oreskovich, who began practicing law in 1989, the year before the first cases were filed. But she wound up representing the largest group of clients, a contingent of 1,500 and 2,000 known as the Berg plaintiffs, named for a plaintiff, Louis Berg.

"The other lawyers called me the 'darling of the people,' " said Oreskovich.

Not only was the case enormous and complex, with formidable opposing counsel representing the best-funded defendant in the world, the plaintiffs' lawyer with the most clients had "never tried a case…and had no experience in personal injury litigation," McDonald said in court papers.

Oreskovich's issues with the other plaintiffs' lawyers began almost immediately.

Seattle solo practitioner Tom Foulds originally brought Oreskovich into the case, hiring her to work for him in 1991."I realized after six weeks that it wasn't going to work out," he said. "But I knew she would connect with clients. She sure did."

Oreskovich continued on her own with the case and was the only plaintiffs' lawyer who refused to dismiss Westinghouse Hanford Co. as a defendant. The company, a subsidiary of Westinghouse Electric Corp., took over running the plant in 1987, at which point virtually no radiation was being released, according to a government environmental report.

The other lawyers saw little point in suing Westinghouse, but Oreskovich forged ahead on her own. In 1994, McDonald dismissed Westinghouse from the case, later describing Oreskovich's efforts as "a waste of the court's time and her clients' money."

In 1995, Oreskovich joined forces with Eymann, a well-known Spokane plaintiffs' lawyer who went on to serve as president of the Washington State Bar Association in 2000. He agreed to fund expert witnesses and other expenses in return for serving as co-counsel to Oreskovich's clients.

Within a few months, the ­arrangement had gone sour. Eymann filed a motion to withdraw from the case in April 1996, citing "irreconcilable differences" with Oreskovich.

"The court is greatly disturbed," McDonald wrote, noting that Oreskovich's prior co-counsel, William Vines, had also withdrawn. "The court indicated it would not tolerate continued disruption in the representation of this plaintiff group."

He demanded a "detailed and exhaustive list" of Eymann's reasons for quitting within seven days. When he got it, he appointed a magistrate judge to investigate the allegations.

McDonald, in court documents, summarized the magistrate judge's 40-page report, saying that it contained more than numerous factual findings of misconduct against Oreskovich (according to a later 9th Circuit opinion, the magistrate made more than 150 findings). Among the issues McDonald cited: billing personal and other client costs to the Hanford account and submitting interrogatory responses that were not reviewed, approved or signed by the clients.

On Nov. 12, 1996, McDonald disqualified her as counsel, writing, "Her incompetency and bad faith conduct threaten the integrity and orderly progress of this litigation."

After five-year investigation, the state bar, in an Oct. 18, 2001 order, dismissed all charges against Oreskovich, though it called her approach "sloppy." "At the end of the day, the bar only affirmed what my 2,000-plus clients had determined for six years consistently — that I had complied with all ethical and competency requirements on their behalf," Oreskovich said.

Meanwhile, Oreskovich's clients were livid. They petitioned the court, the U.S. Court of Appeals for the 9th Circuit, even then-Vice President Al Gore, demanding that Oreskovich stay on. "I am extremely angered that the court system is raping downwinders once again," wrote one client, Kay Sutherland, in a letter Oreskovich submitted to the Washington State Bar Association. "The court system does not have the right to choose my attorney."

About half of her clients dropped out of the litigation. The others retained Eymann or Brian Depew, a partner at Engstrom, Lipscomb & Lack in Los Angeles.

With Oreskovich gone, McDonald temporarily severed the Berg plaintiffs from the rest of the case to give their new lawyers time to catch up.

The others began to drill down into the core of the case. The key issue was causation. There was the generic question: Did Hanford emissions have the capacity to cause the illnesses people complained of? And the specific question: Did each individual plaintiff receive a high enough dose of radiation to cause that person's particular disease?

The plaintiffs believed that to survive summary judgment, they only needed to prove emissions from Hanford could cause the illnesses. After that, the lawyers planned to present evidence about each plaintiff's exposure, disease and other risk factors like diet and family history.

The defendants countered that for claims to move forward, every plaintiff had to show he or she had been exposed to a threshold dose of radiation that statistically "doubled their risk" of harm. Otherwise, the lawyers argued, they couldn't ever prove Hanford caused their illnesses, since the diseases all occur in the general population as well.

Both sides filed summary judgment motions, and in 1998 McDonald issued a 762-page decision that essentially destroyed the plaintiffs' case.

"Proving radiation is 'capable of causing' a disease does not entitle any plaintiff to get his/her case before a jury," he wrote. Instead, he wholeheartedly embraced the defense's theory, writing that the plaintiffs had to show up-front they'd received doses of radiation that doubled their risk of getting sick.

The judge then excluded the testimony of 17 of the plaintiffs' expert witnesses completely or in part as irrelevant. Any plaintiff whose claim relied on one of these experts was removed from the case. In the end, almost no one was left.

"It was a scathing analysis," plaintiffs' lawyer Foulds said. "He went out of his way to dismiss our experts."

The plaintiffs appealed to the 9th Circuit. Two years later, the court resuscitated the case, faulting McDonald for overreaching. "Radiation is capable of causing a broad range of illnesses, even at the lowest doses," wrote then Chief Judge Mary Schroeder for the unanimous panel. "The district court should not have ventured into individual determinations at this stage."

The case, now 11 years old and with little established but the fact that radiation can cause diseases, was sent back to McDonald in 2002.

At this point, it was apparent to the plaintiffs' lawyers that McDonald, who was appointed by President Ronald Reagan in 1985, had little sympathy for their cause. "A true enemy," is how Foulds put it.

Foulds did some digging on the judge and discovered that McDonald owned millions of dollars in local real estate, including orchards and other agricultural lands near Hanford. Of particular interest: the judge's purchase in 1999 of an orchard near Ringold, Wash., where some of the heaviest radioactive contamination occurred. In bank documents, McDonald certified the land was free of contamination, including radiation.

In December 2002, Foulds filed a motion asking McDonald to recuse himself, writing that the properties created a "clear and present acute financial conflict" and "a mind-set that is naturally adverse to any reports of contamination." No co-counsel initially joined in the motion. "Everyone was so scared of a federal judge," Foulds said. "I did it entirely on my own."

But it worked. In a response brimming with anger, McDonald in March 2003 agreed to recuse himself, calling his service "a wasted decade of time and study."

The judge found the timing of the request deeply suspicious, writing that, "[h]owever one looks at it, plaintiffs' recently filed recusal motion follows the rendering of a decision adverse to them." McDonald noted that for the past 17 years on the bench, he had filed annual financial disclosure reports. "Even if the plaintiffs did not bother to review the court's annual financial disclosure, all they had to do was read the newspaper," where he said there had been half a dozen articles about his assets.

Born in 1927 and raised about 75 miles from Hanford in Yakima County, McDonald (who died in 2007) was technically a downwinder himself. He took issue with Foulds' basic premise that a pro-plaintiff ruling might decrease the value of his land. Any land buyer would already be aware that the counties surrounding Hanford "have allegedly been the primary depository of the radioactive ash," he wrote, and noted that area land values have nonetheless increased.

Still, he recused himself. If he failed to do so, he reasoned, he'd just cause more delay, either via an immediate appeal to the 9th Circuit, or as ground for an appeal down the road. And perhaps he was still smarting from a recent reprimand from the circuit — which chastised him for exchanging notes with his clerk referring, for example, to Hispanic people as "greasers" — and dared not risk another shaming.

"Too much of the time in the trial court involved collateral issues," he wrote. "The toll of time on the parties and their rights to have this matter resolved in their lifetime cannot be ignored."

That was eight years ago.

The case was reassigned to Nielsen, a 1991 George H.W. Bush appointee, who faced the daunting task of getting up to speed on proceedings where nearly 1,300 motions, orders and notices had already been filed. "We want to get this case on track, get it moving," he said optimistically in his first meeting with counsel.

He scheduled bellwether trials for 2005, where each side would pick a few plaintiffs, present their cases to a jury and then (in theory) use the verdicts as guideposts to settle the other claims.

The first question was which cases to try. The plaintiffs' lawyers proposed that the proceedings be limited to those involving thyroid disease, which make up approximately two-thirds of the claims. These were the most straightforward to prove — radiation is the only known cause of thyroid cancer, according to the plaintiffs' lawyers. Claims involving other cancers such as bone, liver or stomach, linked to plutonium exposure or contaminants from the Columbia River, would wait.

"Bringing in all these extra different experts, different dose reconstructions, it gets extremely complicated," Roselle said during an August 2003 status conference, according to the court reporter's transcript.

It also gets expensive — more expert witnesses meant more bills to pay, and plaintiffs' lawyers were already deep in the red. Focusing on thyroid cases was one way to conserve resources.

The defense protested that the trials should include a cross-section of diseases, but Nielsen opted to make the proceedings thyroid-only. Since then, the nonthyroid-cancer cases — there are 439 of them — have received scant attention. But last week, the judge said that 30 of these cancer cases may go to trial, with dates and details to be determined in the coming months.

The defense lawyers picked six plaintiffs who had what they said were "representative" claims. Those claims were dismissed on dispositive pretrial motions, court documents say. The other six, the ones picked by the plaintiffs' lawyers, had their cases presented to a Spokane jury in 2005.

Three had hypothyroidism — a condition in which the thyroid is underactive, causing symptoms like fatigue, depression and weight gain. The jury rejected all of their claims, finding Hanford did not cause their illnesses. The verdicts were overturned on appeal in 2007 based on erroneous evidentiary rulings, and the cases have not been retried.

The other three had thyroid cancer. One was awarded $317,000, another, $228,000 after the jury found their cancers were "more likely than not" caused by Hanford.

But huge jury verdicts weren't necessarily expected. That's because the first two plaintiffs had thyroid cancer that was successfully treated — in fact, the five-year survival rate for thyroid cancer is 97%, according to the American Cancer Society.

But the third, Shannon Rhodes, had terminal thyroid cancer that had spread to her lungs. (Rhodes died in May). This was supposed to be the high-value case, the money maker. Eymann in closing arguments asked the jury to award her $20 million to $30 million, according to the trial transcript. "What would a person pay to live another year, to see their daughters, to be held by their husband.…Everybody, who would not pay that, a million a year for those years?" he said.

The jury deadlocked and her case was retried the same year.

The second time around, Rhodes lost. The jury found, 11-1, in favor of the government that emissions from Hanford most likely did not cause her cancer.

How did the plaintiffs' lawyers miscalculate so badly?

In large part, it came down to a single number — the rad, or radiation absorbed dose. The biggest mistake, plaintiffs' lawyers say in retrospect, was agreeing to use rad estimates derived from the government's Hanford Environmental Dose Reconstruction (HEDR). In 1987, the Department of Energy launched the $27 million study under the auspices of the Centers for Disease Control and Prevention but conducted by one of Energy's main Hanford research contractors. Battelle N.W. Computer models factored radioactive emissions plus a person's age, sex, place of residence and diet to come up with his or her rad. Those youngest and closest to Hanford during the years of peak emissions had the highest rads.

By agreeing to use the HEDR model, the plaintiffs streamlined the trials — there was no fighting over how to calculate a rad — and saved money on expert witnesses.

According to HEDR, the first two cancer plaintiffs both had doses of 27 rads. But Rhodes' dose was only 6.9 rads — not enough to convince the jurors that Hanford caused her disease. In fact, one expert testified that by age 30, people in the United States on average have a whole-body dose of up to 10 rads, and a thyroid dose of five to seven rads from sources such as radon, cosmic rays, X-rays and even airplane flights.

"If we had to do it over again, we wouldn't have stipulated to use HEDR," Eymann said. "We did it for trial efficiency, and because we believed at the time that with our experts, we could show that thyroid cancer was caused by doses above one rad."

Foulds was harsher in his assessment. "The decision itself was fundamentally flawed, both scientifically and as a trial tactic," he wrote in an April 2009 motion that highlighted the divisions within the plaintiffs' lawyer ranks. Foulds said he never wanted to use HEDR in the first place, but was "kept out of the loop of discussion on the decision…apparently because [my] opposition was known and on the record," he wrote.

After the bellwether trials, co-counsel decided Foulds was right all along, and that HEDR produced inaccurate, low-ball doses. They now argue the study used incomplete meteorological data plus erroneous assumptions about how the terrain around Hanford influenced the path of the emissions. Also, the model assumed that area cows ate stored feed rather than pasture grass exposed to iodine-131, and failed to reflect the plaintiffs' rural diet of homegrown food, they say.

"Any time there's a range of input data, [HEDR] always chooses the lowest number," said plaintiffs' lawyer Roy Haber, a Eugene, Ore., solo practitioner who represents several hundred plaintiffs.

Going forward, most of the plaintiffs' lawyers are pinning their hopes on an alternate study by Douglas Crawford-Brown, executive director of the Cambridge Centre for Climate Change Mitigation Research, which yields vastly higher rad doses. Using the Crawford-Brown model, the dose received by plaintiff Rhodes, for example, would be 80 rads instead of 6.9, Eymann said.

Scott Blake Harris, who stepped down as general counsel of the Department of Energy in March to head Neustar Inc.'s legal department, finds the numbers improbable.

"In my personal opinion, the Crawford-Brown data doesn't pass the laugh test," he said. "It would predict, literally, thousands of thyroid-cancer cases in the Hanford area. Yet even the plaintiffs' counsel could find only about 250 cases to file."

The Centers for Disease Control and Prevention's 2002 Hanford Thyroid Disease Study, which was mandated by Congress, found rates of thyroid disease near Hanford did not differ from those in the general population. "Subsequent scientific technical reviews had substantial criticism in the way in which the study was conducted and the results were reached," Eymann said.

A later study found men who grew up near Hanford had higher rates of thyroid disease — but didn't link this to iodine-131.

The bellwether trials have proven to be of little use in settling the case. During Harris' two-year tenure as general counsel, the two sides mediated about 50 thyroid-cancer cases, and the government made settlement offers for the first time to all the thyroid-cancer plaintiffs. "We tried to settle every case we thought was even arguably meritorious, and even made nominal offers in cases that weren't — simply because we thought bringing these cases to a close would be in the public interest," Harris said.

Those with the highest rads were offered up to $150,000 — on par with the Department of Labor's standard payment to Hanford workers who later contracted certain cancers. Plaintiffs with the lowest rads — some as low as 0.5 — received offers in February of $10,000.

"We really had to think seriously about making the $10,000 offers," said a senior official at the Department of Energy who asked not to be identified by name because the litigation is pending. "For taxpayers to lay out any significant amount of money, plaintiffs had to lay out some proof that what they experienced was actually caused by Hanford." About 30 to 40 plaintiffs accepted the offers, he said.

It's not just the taxpayers' money. Any settlement payments come from the Department of Energy's budget, not the U.S. Treasury. "We don't have all that money sitting around," the official said. "We have to find it or get it appropriated." That feat will only be more difficult going forward, given the federal budget crisis.

The offers were not well-received by the plaintiffs' lawyers. "The value of the case [to the government] is determined by the dose, not by how much someone suffered," said Depew of Engstrom Lipscomb, who represents about 400 plaintiffs. "They refuse to compensate people based on the injury. Whether someone received an extra rad or two doesn't make their cancer worse or better."

A few of his clients accepted the offers, he said, but only out of desperation. One woman was dying, while another was homeless and used the money to buy a trailer.

Plaintiffs' lawyers in court papers say further settlement discussions are going nowhere because the defendants "cling to the flawed HEDR model" in deciding the value of the cases. To the government, low rads equal a low offer, but plaintiffs assert the dose estimates themselves are not valid.

Harris responded, "There is no way that the government can settle these cases based on the assumption that all of the science accepted by the courts so far is wrong."

The plaintiffs hope to change that. In May, they filed a longshot motion asking the court to exclude HEDR rad doses from evidence going forward on the ground that the data would mislead a jury. A hearing is set in September,

A new trial is scheduled for April 30, 2012, for 20 hypothyroid plaintiffs chosen by lottery, and determining their rads is likely to be one of the most critical issues.

Last week, the judge also ordered trials for 32 to 33 thyroid-cancer cases and 50 cases involving thyroid nodules — solid or fluid-filled lumps that form within the thyroid — with dates to be determined.

But for one plaintiff, Deborah Clark, all the wrangling over doses and rads didn't change the fact that she was dying.

Clark, who grew up in eastern Oregon drinking raw cow's milk, had thyroid cancer that spread to her lungs. Eymann claimed her radiation dose was 35 rads and wanted $2 million to settle her case. The government said it was less than one rad and offered $10,000.

Last year, Eymann made an impassioned plea to the court to grant her an expedited trial, even though it would throw off the schedule for the hypothyroid trials. "It is Mrs. Clark's desire that she 'have her day in court' prior to her premature death," he wrote. "The issue is her right to justice versus efficiency concerns." And he argued that granting her a trial within five months could help resolve questions over HEDR and Crawford-Brown that have brought the rest of the case to such an impasse.

But lawyers from Kirkland & Ellis were having none of it. The plaintiffs are "trying to cherry pick claims and derail a carefully conceived process," partner Bradley Weidenhammer wrote. "Mrs. Clark is not unique. Time is of the essence for a large number of elderly and ill plaintiffs in this case."

"There is no evidence suggesting she will die within one year, as opposed to two or three or 10 years," he continued in a motion filed on Feb. 10, 2010. He was right. She didn't die within a year. It was a year and two days, on Feb. 1 2, 2011.

The defense even opposed a bid to conduct an expedited deposition of Clark's mother, then 79 and in failing health, who was the only person who could testify about her daughter's milk intake as a child. Weidenhammer wrote that it was "counter-productive to pursue discovery as to plaintiffs who are not yet scheduled for trial" — even, presumably, if it meant the evidence dies with them. Nielsen denied the motion for speedy trial, but ordered the expedited deposition of Clark's mother.

It's increasingly clear for all the plaintiffs that time is running out. As of last year, more than 800 were older than 70, and 174 were older than 90. Eymann reports that 123 of his clients — about 30% of them — have already died.

Looking ahead, it's hard to see how the litigation will end. At the moment, neither side seems inclined to budge on settlement amounts, but the alternative — trying each case, one by one, before a jury — could take decades.

If you ask Van Wart, it's the plaintiffs' lawyers' fault that the case has dragged on for so long. "The problem for the plaintiffs is that it costs more for them to take the cases to trial than they can recover," he said. "The plaintiffs' lawyers are standing in the way of resolution because of their financial predicament. They're hoping somehow, someone will bail them out and give them a big basket of money."

The plaintiffs' lawyers are equally quick to blame the defendants for refusing to make what they consider to be fair offers. "No one has big dreams. No [plaintiffs'] lawyer is going to see a dime of profit off this case," Depew said. "The only one making money is Kirkland & Ellis. I just want something for my clients."

The downwinders are left stranded in the middle. People with valid claims have not received the "full and prompt compensation" that the Price-Anderson Act promised them nor have they had a chance for a jury to evaluate their cases. Consider Sharon Bentz, who is 62. Born in Spokane and raised in the little town of Elmira, Ore., she's one of the plaintiffs who showed up for the status conference in late April. In recent years, she's been terribly ill — a carcinoid tumor in her appendix, a heart attack, diabetes, glaucoma, a hysterectomy, hives covering her body — and thyroid disease. Such an array of ailments doesn't make much sense to her, except maybe Hanford has something to do with it.

Whether she'll ever have her day in court, with expert witnesses opining on the cause of her illnesses, is an open question, one she is ill-equipped to answer. As she is sitting outside the Spokane courtroom, on the bench next to Carole Means, Eymann walks by. Neither woman recognizes him. They've never met their lawyer.

"It's been such a burden for my family and friends. I've been so ill," Bentz said. "I know the attorneys are doing their jobs." But after so many years of waiting, she added, "I just get frustrated."