JACKSON, District Judge.

Sixteen years ago Carolyn Bayless began to suffer from a mysterious debilitating illness. As her condition deteriorated over the years that followed, she doggedly sought to learn what caused (and how to treat) her illness. Finally, in 2008, convinced that she was the victim of exposure to nerve gas emitted by an Army testing facility, she filed a claim under the Federal Tort Claims Act. When this lawsuit followed in 2009, the Army responded that she knew of her claim by at least 2005 and had waited too long to assert it. The district court agreed and granted summary judgment dismissing the case. We conclude that under the unusual circumstances presented here, the period of limitation did not accrue until February 2007. Therefore, exercising jurisdiction pursuant to 28 U.S.C. § 1291, we reverse.

I. BACKGROUND

A. Factual Background

In 1997, after graduating from college with a conservation biology degree, Ms. Bayless began a seasonal position as a range technician with the Utah Division of Wildlife Resources. As a result, she traveled to remote locations around the state to conduct various wildlife studies, including one location less than ten miles away from the Dugway Proving Grounds ("Dugway") and another location within two miles of Tooele Chemical Agent Disposal Facility ("Tooele"). Unbeknownst to Ms. Bayless during the time in which she worked near them, both Dugway and Tooele were United States Army sites then conducting chemical and biological weapons testing.

In October 1997, about a month after completing her seasonal position with the Utah Division of Wildlife Resources, Ms. Bayless began to experience episodic lip numbness and blurred vision, which she attributed at that time to sitting in front of the computer for numerous hours a day. After the symptoms failed to dissipate, however, Ms. Bayless sought medical evaluation. In February 1998 she saw an ear, nose, and throat specialist who performed an MRI, which returned normal results. Shortly afterwards and notwithstanding those normal results, Ms. Bayless experienced numbness through her entire left side following cleaning her house with chemicals. Ms. Bayless thus began a long and arduous search for an answer to her worsening medical condition.

In May 1998, Ms. Bayless sought evaluation from a neurologist, Dr. Christopher Reynolds, whose tests likewise revealed normal results. The numbness meanwhile spread to Ms. Bayless' right foot and right arm. After severe vertigo sent Ms. Bayless to the emergency room in July 1998, she discontinued her birth control pills upon her doctor's advisement that she suffered a transient ischemic attack.1 Later in the same month, Ms. Bayless sought out a second neurologist, Dr. Dennis Thoen, whose tests again showed normal results.

In January 1999, however, Dr. Thoen began to suspect that Ms. Bayless' symptoms were caused by multiple sclerosis ("MS"). After a series of referrals, in February 1999 a neurologist and MS specialist, Dr. John Rose, formally diagnosed Ms. Bayless with MS and prescribed medication accordingly. Ms. Bayless' condition, however, only continued to worsen, and she discontinued the MS medication in August 1999.

Ms. Bayless' condition thereafter seemed to improve, and she became pregnant in January 2000. Unfortunately, her improvement proved to be temporary. In March 2000 Ms. Bayless' health took a dramatic turn for the worse when she suffered a miscarriage and underwent an emergency dilation and curettage. Two days later she began to experience severe difficulty walking, and she lost control of her fine motor skills. She became unable to care for, even to feed, herself. Ms. Bayless described it as going from functional to nonfunctional.

Still under the theory that this sudden deterioration was related to MS, Dr. Rose prescribed oral steroids on March 24, 2000. The steroids only worsened Ms. Bayless' condition. Despite what was happening, however, another MRI performed on July 3, 2000 yet again showed normal readings.

In July and August 2000, Ms. Bayless traveled to Chicago to another MS specialist, Gastone Celesia, M.D., who determined that she had been improperly diagnosed with MS. Instead, Dr. Celesia and another specialist, Dr. Tony Fletcher, both suspected Ms. Bayless' symptoms were psychosomatic.

Veering Ms. Bayless in another direction in October 2000, Utah physician Judith Moore, D.O., diagnosed Ms. Bayless with a chronic Epstein-Barr viral infection. Dr. Moore prescribed vitamins and supplements to treat the infection. But a month later Dr. Dean Wingerchuk, a neurologist at the Mayo Clinic in Arizona, seconded Drs. Celesia's and Fletcher's conclusion that Ms. Bayless did not suffer from MS and advised her that the psychosomatic symptoms would disappear without medical treatment. Hoping that her symptoms would dissipate as these doctors had counseled her, Ms. Bayless stopped seeking medical treatment. In August 2001 she gave birth to a daughter in Colorado. Although Ms. Bayless continued to suffer from the same neurological problems, she did not seek treatment for another year after her daughter's birth.

In October 2002, Ms. Bayless, now having discarded the MS theory and suspecting that her symptoms might be related to neck issues, began treatment with a series of chiropractors. On her intake forms with one of the chiropractors, Dr. Beau Maudlin, Ms. Bayless mentioned her work in southern Utah and raised concerns over nuclear testing and mining tailings in the area. Dr. Maudlin referred Ms. Bayless to another chiropractor, Susan Rector, at the end of March 2003. Dr. Rector raised her suspicion that Ms. Bayless' symptoms might be caused by heavy metal or mercury poisoning. Ms. Bayless was then referred to a dentist to remove her mercury dental fillings, but these detoxification treatments were again unfruitful in resolving her symptoms.

Subsequently, still in 2003, Ms. Bayless was referred by her dentist to a clinical nutritionist, Sam Queen. On the "Possible Exposure" portion of her patient intake form, Ms. Bayless wrote, "I think I may have been exposed to uranium in the soil in southern Utah when I worked for the Division of Wildlife for 6 months collecting plant and soil samples." App. to Appellant Br. 64. Mr. Queen performed blood work analysis and informed her that he suspected her symptoms were caused by "organophosphate pesticide poisoning." Id. Ms. Bayless assumed this meant she had been exposed to lawn pesticides and, while she did not pursue an additional investigation into the diagnosis, she began Mr. Queen's detoxification regimen. However, Ms. Bayless stopped that regimen in December 2004 on the recommendation of her physician at the time, Dr. Dennis Remington, who noted that she suffered from an apparent chronic renal failure.

In early 2005, Ms. Bayless read a magazine article by a woman named Pauly who had experienced similar symptoms. When Ms. Bayless spoke with Pauly on the telephone, she learned that Pauly developed her neurological symptoms after visiting an Army base. Pauly suspected her own symptoms were caused by biological weapons testing at the base, and she referred Ms. Bayless to Dr. Garth Nicolson, a research professor at the Institute of Molecular Medicine in San Diego who studies biological weapons.

Dr. Nicolson suggested to Ms. Bayless various methods of detecting exposure to biological weapons, including a polymerase chain reaction ("PCR") test. Ms. Bayless did not obtain a PCR test at that time due to "debates on whether the PCR is accurate or not." Id. at 73. Notably, Ms. Bayless did not discuss nerve agents with Dr. Nicolson, only biological agents and the related conditions caused by mycoplasma fermentans and mycoplasma incognitus.

In early 2005, after speaking again with Pauly, Ms. Bayless began to research biological and chemical weapons testing. Upon discovering the existence of the Army sites at Dugway and Tooele, Ms. Bayless again consulted Mr. Queen regarding any possible exposure to biological and chemical weapons. On February 1, 2005, Mr. Queen wrote to Ms. Bayless that normally her lack of improvement from pesticide detoxification "may be because of exposure to various toxins." Id. at 66, 107. Mr. Queen suggested that a "secondary toxin ... due to nerve agents, etc.," might be causing the lack of improvement in her case. Id. Mr. Queen wrote that "the Institute of Orthomolecular Medicine ... would probably have a way to test for this," and that "the information may be very useful." Id.

Ms. Bayless did not follow up with this Institute. Instead, on February 9, 2005 Ms. Bayless raised the issue with Dr. Remington who recorded that he discussed "new ideas" with Ms. Bayless about her work history adjacent to Dugway and Tooele. Id. at 68, 97. Dr. Remington's notes do not record — and Ms. Bayless does not remember — whether he recommended any treatment or follow-up after this discussion.

Ms. Bayless also began seeing Dr. Robert Moody, another physician in Utah, in early 2005. Among other things Ms. Bayless provided contact information for Major Craig Lynch at Dugway, who was a contact provided by Dr. Nicolson from the Institute of Molecular Medicine, to Dr. Moody. Id. at 111. In her note to Dr. Moody, Ms. Bayless mentioned that Dr. Nicolson had purportedly helped Major Lynch with his Gulf War Illness. However, neither Ms. Bayless nor Dr. Moody followed up by contacting Major Lynch.

In an undated letter from early 2005 to Mr. Queen, Ms. Bayless commented on her mother's medical history while she was pregnant with Ms. Bayless. Ms. Bayless remembered that her mother took two anticonvulsant drugs during the pregnancy, and she also remembered some neurological problems in her childhood. She wrote, "My hunch is that these two drugs caused a weakness in my central nervous system/brain so that when I was exposed to a neurotoxin near Dugway it explains why it affected me so badly." Id. at 109.

Ms. Bayless also wrote that she had found a doctor2 who believes that she "was exposed to a neurotoxin out at Dugway and is willing to pursue it and call out there." This doctor, she claimed, also believed that the amount of acetylcholine in Ms. Bayless' muscles was deficient. Ms. Bayless noted that she had read "that organophosphate toxicity (like what they believe may be affecting some Gulf War vets) damages the acetylcholinesterase enzyme so there is a buildup of acetylcholine in the nerve endings which causes neuromuscular problems." Id. at 68-69. Mr. Queen wrote in his reply — also undated — that this is "possible," and that treatment with huperzine would eliminate any acetylcholinesterase inhibitors.3

On March 15, 2005, Dr. Moody recorded in Ms. Bayless' history that "[a]ll [signs and symptoms] preceded by working at forestry service near Dugway [with] exposure to Seran [sic], Brucelloses [sic], multiple infectious disease." Id. at 74, 114. On April 12, 2005, Dr. Moody also recorded that Ms. Bayless "[m]ay have been exposed to nerve gas being destroyed [at] Toela [sic] Army." Id. at 73, 113.

Meanwhile, also around April 2005, Ms. Bayless met with Dr. Rebecca Levine, a family practitioner, on yet another theory that her symptoms might be related to seizure activity. After hospital testing revealing negative results, Ms. Bayless crossed off seizure activity as a potential cause.

At some point in early 2005 during her own research, Ms. Bayless had learned of a man named Gary Harris who had worked at Dugway in 1996 and who had suspicions that he had been exposed to nerve agents. Mr. Harris shared similar symptoms to those experienced by Ms. Bayless, including "brain fog, numbness, muscle weakness, and chronic fatigue." Id. at 71, 116. Mr. Harris informed Ms. Bayless that "the government was shooting munitions filled with VX and GB nerve agents into the southwest corner of Dugway and lost a lot of them outside the borders." Id. at 116. He also informed Ms. Bayless that she could have been exposed if she had been downwind from the open burns and detonations that the government performed during that time.

On May 3, 2005, Ms. Bayless saw Dr. Moody again, who recorded in his notes: "Exposure to possible Seran/VX [sic]. [1996-1997] shot munitions into SW corner." Id. at 113. On the same day, Ms. Bayless asked Dr. Moody to test her acetylcholine levels.

On May 4, 2005, Ms. Bayless, citing her own research and the information she learned from Mr. Harris about Dugway, wrote in an email to Mr. Queen: "I now am pretty sure that my symptoms were caused by low-level exposure to nerve agents, probably VX and GB (sarin nerve gas)." Id. at 116. Ms. Bayless also wrote that she found out that employees in the Tooele facility, which operated "the largest nerve gas incinerator in the world," have also suffered neurological problems. Id.

Nevertheless, around the same time in early May 2005,4 Ms. Bayless received negative results from Dr. Moody's acetylcholine test. Later between July and October of 2005, Ms. Bayless was also treated by Mr. Queen with huperzine, a Spanish club moss that he believed should have neutralized effects of sarin nerve gas by targeting inhibitors of acetylcholinesterase enzymes and thereby restored Ms. Bayless' acetylcholine levels. After three days of treatment, however, Ms. Bayless' condition not only failed to improve but it actually worsened. Mr. Queen therefore discontinued any theory that an organophosphate toxicity would be linked to sarin, and Ms. Bayless ended her search for a chemical weapons link.

In early 2006, Ms. Bayless began seeing another nutritionist in Utah, Pat Montague. Ms. Bayless, having abandoned the possibility that her condition was related to Dugway or Tooele, worked with Ms. Montague's suspicion that a Candida infection in Ms. Bayless' intestinal tract was to blame. Ms. Bayless adhered to a strict diet to treat this until October 2006 — but again to no avail.

Returning again to biological weapons, Ms. Bayless visited another Utah family medicine specialist, Dr. Todd Mangum, in October or November 2006. Dr. Mangum administered the PCR test for biological weapons, the test previously recommended by Nicolson in 2005. The test results for mycoplasma fermentans and mycoplasma incognitus returned negative in December 2006.

Finally, Ms. Bayless sought out Dr. William Rea, a specialist in neurotoxicity at the Environmental Health Center in Dallas, Texas. During her first visit with Dr. Rea on February 5, 2007, Ms. Bayless told him about her work in Utah near a nerve gas incinerator. Dr. Rea recorded in his notes that Ms. Bayless "strongly suspects that she was exposed to sarin nerve gas at the Dugway Proving Ground." Id. at 118. Dr. Rea then administered, among many other scans and tests, a cholinesterase test. For the first time, Ms. Bayless received positive test results showing low cholinesterase levels — an opposite result from the acetylcholine test performed by Dr. Moody in May 2005 — now indicating that she suffered from what Dr. Rea called an "organophosphate pesticide toxicity." Id. at 76; see also id. at 120.

Dr. Rea diagnosed Ms. Bayless in a June 21, 2007 report with toxic encephalopathy, toxic effect of molds, toxic effect of pesticides, toxic effect of metals, chronic fatigue, fibromyalgia, autonomic nervous system dysfunction, vasculitis, and immune deregulations. Id. at 124. He concluded that Ms. Bayless suffered from moderate neurotoxicity while noting that her "visual and respiratory symptoms ... are delayed sequelae in Sarin exposure." Id. at 126. Dr. Rea wrote that he "firmly believe[d] that in all medical probability this patient's incapacitation is a result of her exposure to pesticides, heavy metals, molds, and mycotoxins in the workplace." Id. at 127.

B. Procedural History

Ms. Bayless filed an administrative claim on January 31, 2008, alleging that the activities of United States Army on its Dugway facility caused her to sustain permanent neurological and other injuries. Id. at 129-34. On May 29, 2009, Ms. Bayless filed a complaint under the FTCA in the United States District Court for the District of Utah.

On December 2, 2011, the government moved for summary judgment to dismiss Ms. Bayless' complaint for lack of subject-matter jurisdiction. The government argued that Ms. Bayless failed to present her administrative claim within two years of the accrual of that claim pursuant to 28 U.S.C. § 2401(b). After hearing argument on April 19, 2012, the district court, granted the motion on May 17, 2012. The district court concluded that Ms. Bayless acquired enough knowledge by May 2005 for her claim to begin to accrue and also that her claim was not protected by the doctrine of equitable tolling. Ms. Bayless appealed.

II. DISCUSSION

"Absent a waiver, sovereign immunity shields the Federal Government and its agencies from suit." Dahl v. United States, 319 F.3d 1226, 1228 (10th Cir. 2003) (quoting Fed. Deposit Ins. Corp. v. Meyer, 510 U.S. 471, 475, 114 S.Ct. 996, 127 L.Ed.2d 308 (1994)). The FTCA waives the United States' sovereign immunity from tort claims and makes it liable "in the same manner and to the same extent as a private individual under like circumstances." 28 U.S.C. § 2674. A tort claim against the government, however, is "forever barred unless it is presented in writing to the appropriate Federal agency within two years after such claim accrues." 28 U.S.C. § 2401(b). Thus as a threshold matter, timeliness "is one of the conditions of the government's waiver of sovereign immunity under the FTCA, and [a federal] court lacks subject matter jurisdiction to proceed under the FTCA if a plaintiff fails to satisfy the FTCA's timing requirements set forth in § 2401(b)." Harvey v. United States, 685 F.3d 939, 947 (10th Cir.2012) (quoting Franklin Sav. Corp. v. United States, 385 F.3d 1279, 1287 (10th Cir. 2004)).

Whether a plaintiff has "file[d] an FTCA claim within the two-year statute of limitations period is a matter we review de novo." Plaza Speedway Inc. v. United States, 311 F.3d 1262, 1266 (10th Cir.2002). In construing when a claim accrues within the meaning of § 2401(b), we seek to interpret the section consistent with congressional intent. Cannon v. United States, 338 F.3d 1183, 1190 (10th Cir.2003) (citing United States v. Kubrick, 444 U.S. 111, 117-18, 100 S.Ct. 352, 62 L.Ed.2d 259 (1979)).

"Both § 2401(b) and its legislative history are silent as to the meaning of `accrues.'" Plaza Speedway, 311 F.3d at 1267 (quoting Arvayo v. United States, 766 F.2d 1416, 1419 (10th Cir.1985)). Section 2401(b) serves "to require the reasonably diligent presentation of tort claims against the government," and like other statutes of limitations, it reflects the "legislative judgment that it is unjust to fail to put the adversary on notice to defend within a specified period of time, and that the right to be free of stale claims in time comes to prevail over the right to prosecute them." Arvayo, 766 F.2d at 1418-19 (quoting Kubrick, 444 U.S. at 123, 100 S.Ct. 352) (internal quotation marks omitted).

The general accrual rule for FTCA claims is the "injury-occurrence rule," where the tort claim accrues on the date of injury. Plaza Speedway, 311 F.3d at 1267. In exceptional cases, however, the "discovery rule" applies to "protect plaintiffs who are blamelessly unaware of their claim because the injury has not yet manifested itself or because the facts establishing a causal link between the injury and the medical malpractice are in the control of the tortfeasor or otherwise not evident." Plaza Speedway, 311 F.3d at 1267 (quoting Diaz v. United States, 165 F.3d 1337, 1339 (11th Cir.1999)) (internal quotation marks omitted). In these cases, the date of accrual is when that reasonably diligent plaintiff knows or should have known of both the existence of and cause of the injury. Id.; Cannon, 338 F.3d at 1190.

We hardly question that this is an exceptional case where the discovery rule applies. Ms. Bayless' injury occurred in 1997 when she worked as a range technician near the Dugway and Tooele Army bases in Utah, whose apparently undisclosed weapons testing exposed her to sarin or other nerve agents. At the time of that occurrence, Ms. Bayless had no reason to suspect the nature of the injury about to befall her, let alone its cause. As such, "contrary to Congressional intent, strict application of the occurrence rule here would deny [Ms. Bayless her] right to compensation under the FTCA." Plaza Speedway, 311 F.3d at 1268.

Starting with the episodic numbness and blurred vision in October 1997, and as each symptom of her injury revealed itself, Ms. Bayless would develop knowledge of an ever-worsening injury. But, at least initially, Ms. Bayless had no knowledge of Dugway or Tooele to connect her time spent near them to her injury. From the time she first saw a doctor in February 1998 after her initial symptoms did not dissipate, however, Ms. Bayless persevered in her search for an answer despite dead end after dead end.

The question before us then is at what point during this long search did Ms. Bayless know or should she have known of the causal link between her condition and the Army's activities. Ms. Bayless contends on appeal that the two-year limitations period began to run no earlier than February 2007 when she received the positive cholinesterase test results from Dr. Rea. That theory, says the government, ignores the teaching of Plaza Speedway.

In Plaza Speedway we applied the discovery rule to an FTCA toxic tort case where the Army's activities at an airfield adjacent to the claimants' racetrack contaminated the property's groundwater and soil. Speedway's owners purchased the property in 1989 with knowledge of the Army's chemical operations but, despite this knowledge, did not test its wells or seek any environmental assessment. In April 1993, the Kansas Department of Health and Environment ("KDHE") tested the well water following an application from the owners for public use of the property's water. Four months later an environmental geologist from the remedial section of KDHE telephoned the property owners and discussed potential chemical contamination on the Speedway premises. Id. That geologist followed up with an October 18, 1993 letter indicating that hazardous substances were present in the groundwater and soil. The owners filed their administrative claim on October 13, 1995.

We held there that the geologist's phone call in August 1993 started the clock for the statute of limitations. 311 F.3d at 1271. In doing so we emphasized three antecedent facts that, when combined with that telephone call, should have given the owners "reason to suspect the source might have been the neighboring property" and then to initiate inquiry into "any possible harm": (1) the owners knew that contamination testing by the KDHE occurred in April 1993; (2) the owners knew at the time of purchase of the property that the Army had been testing various chemicals on the adjacent airfield; and (3) the property had no other neighbors than the Army airfield. Id. at 1270-71.

The facts in Ms. Bayless' case are rather different. Unlike the owners in Plaza Speedway whose options for suspicion could have pointed only to its sole neighbor, Ms. Bayless had more than a handful of potential "suspects" wreaking havoc on her body. And unlike the owners in Plaza Speedway, Ms. Bayless did not sit back idly, with knowledge of the government's activities and without taking any action to inquire into the source of her injury.

The government stresses that Ms. Bayless failed to pursue certain leads, including Dr. Nicolson's suggestion to take a PCR test, Dr. Nicolson's reference to Gulf War Illness and Major Lynch, and Mr. Queen's mention of the Institute of Orthomolecular Medicine.5 Ms. Bayless, however, made a conscious choice at the time against taking the PCR test because its accuracy was debated — indeed, a PCR test administered in 2006 by Dr. Mangum returned no fruitful results.

As for the other references, the test for accrual does not require that a plaintiff pursue every possible clue to its end in order to be deemed a reasonably diligent plaintiff. See Arvayo, 766 F.2d at 1422 (question of reasonable diligence is an objective one). In fact, within a decade Ms. Bayless had seen as many medical professionals as others might see in a lifetime, and she traveled throughout the country to do so. On advice from those physicians, neurologists, chiropractors, nutritionists, and other professionals, Ms. Bayless passed through theories on ischemic attacks, multiple sclerosis, an Epstein-Barr infection, a neck-related condition, heavy metal poisoning, uranium exposure, a seizure-related condition, a systemic yeast infection, and was even told by more than one doctor that her symptoms were entirely psychosomatic. To hold that Ms. Bayless must have followed every word of advice from every professional to be deemed reasonably diligent would change her legal duty into an absurd herculean task.

Despite her diligence, Ms. Bayless remained unaware of even the existence of the Army sites at Dugway and Tooele until early 2005. According to the government, upon this discovery in 2005, Ms. Bayless became not only armed with the critical facts regarding the activity that occurred Dugway and Tooele, but she had already received direction from her doctors that should have connected the cause and injury. The government relies on the following pieces of evidence that Ms. Bayless had by that time: first, Ms. Bayless' research on Dugway and Tooele; second, Ms. Bayless' conversations with "Pauly" and Mr. Harris about suspicions of their own exposure to biological or chemical weapons; third, Mr. Queen's prior "organophosphate pesticide toxicity" diagnosis in 20036 as well as his intimations at a "secondary toxin ... due to nerve agents, etc." in his February 1, 2005 email; and finally, most significantly, Ms. Bayless' own statement in her May 4 letter to Mr. Queen that she was "pretty sure" sarin poisoning was to blame for her condition.

The crux of this case, however, lies in whether a legal distinction exists between one's own suspicion that something is a cause and the requisite objective knowledge that begins the accrual clock. An example from the Seventh Circuit makes the point that subjective suspicion is not necessarily enough.

In Stoleson v. United States, 629 F.2d 1265(7th Cir.1980), the plaintiff, an employee in an Army munitions plant, suffered a heart attack in 1968 and suspected that it was related to her workplace exposure to nitroglycerin. Her condition was diagnosed, however, as a myocardial infarction caused by vascular spasm triggered by temporary withdrawal from nitroglycerin; and her treating physician told her that exposure to nitroglycerin was not the cause. She nevertheless requested (unsuccessfully) a transfer to a work area free from nitroglycerin, and she continued to experience periodic angina attacks until her employment was terminated in 1971.

Meanwhile, her suspicion of a nitroglycerin connection continued. In the spring of 1969 she read an article suggesting that sudden withdrawal from nitroglycerin can cause angina chest pains. Later that year an occupation safety inspector told her that he believed her heart problems were caused by exposure to nitroglycerin, though he was unaware of any medically recognized causal relation. But her personal physicians confirmed the previous assessment, and the facility continued to assure her that her suspicions were groundless.

In April 1971, a cardiologist examined her and concluded that her cardiovascular problems were related to nitroglycerin exposure. Later he went further and, based on the plaintiff's case and other case histories, published an article documenting for the first time the relationship between angina and chronic exposure to nitroglycerin. Notwithstanding the cardiologist's opinion, however, the facility's doctor was unconvinced, and the plaintiff ultimately was discharged as unable to work.

The plaintiff filed her administrative claim on August 16, 1972, and after striking out there, filed suit under the FTCA. In granting the government's motion to dismiss the district court found, among other things, that if the discovery rule applied, her claim accrued in November 1969 — more than two years before she filed her administrative claim — when she read the article and spoke to the safety inspector.

The Seventh Circuit reversed. "A layman's subjective belief, regardless of its sincerity or ultimate vindication, is patently inadequate to go to the trier of fact." Stoleson, 629 F.2d at 1270. Because medical science did not at that time recognize the causal connection, "[t]o fix the time of accrual at this time would provide [plaintiff] with nothing more than a delusive remedy." Id. Only in 1971, when the cardiologist documented the relationship between occupational exposure to nitroglycerin and angina, informed the plaintiff of his findings, and also informed her of their relevance to her medical problems did her suspicion ripen into knowledge sufficient to trigger the running of the two-year limitations period. Id. at 1270-71; see also Rosales v. United States, 824 F.2d 799, 805 (9th Cir.1987) ("Ordinarily, a plaintiff cannot be expected to discover the general medical cause of his injury even before the doctors themselves are able to do so."); Nicolazzo v. United States, 786 F.2d 454, 456 (1st Cir.1986) ("It was only when [the plaintiff] received a correct diagnosis ... that the factual predicate of his injury ... became known to him."); Osborn v. United States, 918 F.2d 724, 733 (8th Cir.1990) (rejecting the government's argument that the plaintiff's mother should have known the cause of plaintiff's seizure "when the doctors, including specialists .. . had not yet reached a conclusion"); Harrison v. United States, 708 F.2d 1023, 1028 (5th Cir.1983) (statute tolled because "[i]t would be unreasonable to hold [a plaintiff] to a higher degree of medical competence and understanding than the many medical experts she consulted").

Stoleson is distinguishable. There is no indication in the facts here that medical science did not recognize a connection between exposure to nerve agents and symptoms such as those of Ms. Bayless before she saw Dr. Rea in 2007. The point, however, is that what Ms. Bayless had before that time was a lay person's suspicion. While in some circumstances that might be sufficient to trigger the statute, it cannot do so here when the context is considered.

In early 2005, Ms. Bayless did, of course, know of the existence of the Army bases. She suspected, at times, that their activities caused her injuries. But when she followed up on her suspicions with various professionals she received, if anything, objective medical results to the contrary. And it was not just one doctor who turned Ms. Bayless away. She received three empirical results from three different doctors, all evincing what appeared to be the objective truth that Dugway and Tooele were not to blame. First there was a negative acetylcholine test in early May 2005 by Dr. Moody, notably administered around the same time that she wrote she was "pretty sure" that sarin gas exposure was the cause of her condition. Next, Ms. Bayless was treated with huperzine between July and October 2005, but to no avail despite Mr. Queen's assurance that it would neutralize the effects of sarin nerve gas. Last there was the negative PCR test for biological weapons exposure conducted by Dr. Mangum in December 2006.

The objective indicia seemed to point to a conclusive "no" for chemical and biological weapons exposure, and Ms. Bayless wholly dropped the idea of biological or chemical agent exposure.7 Instead she pursued treatment with Dr. Montague throughout 2006 on a theory that she suffered from a Candida infection. Once again, the new theory failed, and her evidence is that it was not until her consultation with Dr. Rea and the ensuing cholinesterase test8 in February 2007 that she knew that her suspicions were valid.

The government posits that if we were to conclude that Ms. Bayless' claim did not accrue until she found Dr. Rea, it could lead the statute of limitations in many cases to "be deferred indefinitely while the prospective claimant searched for some physician willing to connect her condition with government activity." Thus, "the weaker the plaintiff's claim, the longer the accrual of the claim would likely be deferred." Appellee Br. 34. But merely because a cause is more difficult to determine medically does not make that claim legally weaker. The purpose of the discovery rule inquiry is to recompense the reasonable diligence of an enduring plaintiff despite a scientifically or medically challenging claim. One might equally worry that under the government's view, conspiracy theorists would run to the nearest courthouse to file futile suits for every injury they suspect had been caused by the government's activities — even where that suspicion is based on anecdotes from strangers and Internet gossip, even where that suspicion is repeatedly refuted by objective, medical documentation to the contrary. The discovery rule for accrual should not serve to reward those who tilt at windmills yet snub reasonable and diligent plaintiffs.

To be clear, our holding today does not upend precedent in this Circuit or others that "compelling" or "certain" proof of a cause is not a requirement before accrual may begin. See, e.g., Kronisch v. United States, 150 F.3d 112, 123 (2d Cir.1998); Nemmers v. United States, 795 F.2d 628, 631 (7th Cir.1986). Nor does a plaintiff in every case need medical or scientific confirmation of a cause before the statute of limitation begins. We only opine with respect to the facts of Ms. Bayless' case that, after confronting demonstrable evidence debunking her own suspicions, she cannot be charged as a matter of law with knowledge under the discovery rule. We hold only that, in the circumstances presented by the parties' respective summary judgment filings, that Ms. Bayless' administrative claim was timely filed.

Alternatively, Ms. Bayless argues that the district court erred as a matter of law in holding that the doctrine of equitable tolling did not apply to FTCA claims. Because we reverse on grounds that Ms. Bayless' claims accrued in 2007 less than two years before she filed her administrative claim, we need not reach whether the district court erred in holding that the doctrine of equitable estoppel did not apply. Accordingly, we REVERSE the district court's dismissal for lack of subject-matter jurisdiction and REMAND for further proceedings consistent with this opinion.9

HARTZ, Circuit Judge, concurring and dissenting:

I agree that the summary judgment in favor of the government must be set aside. Reading the evidence in the record before us in the light most favorable to Ms. Bayless, one could say that a person in her situation acting with reasonable diligence would not have discovered the cause of her ailments more than two years before she filed her FTCA claim.

But the majority opinion does more than set aside the government's summary judgment. It grants a (partial) summary judgment to Ms. Bayless on the limitations-period issue. It holds as a matter of law that she filed her claim in a timely manner. It so holds even though Ms. Bayless did not seek such a ruling in district court1 and did not argue in support of such a ruling in this court.2 The 2010 amendments to Fed.R.Civ.P. 56 now prohibit the district court from sua sponte granting summary judgment (and presumably partial summary judgment as well) without "giving notice and a reasonable time to respond." Id. Rule 56(f). The majority apparently do not believe that a similar principle should apply on appeal. But why not give the government the same opportunity it would have in district court? Indeed, why not follow our customary practice of deciding only the issues presented to us and remanding further matters to the district court? Because I think the relief granted by the majority opinion is inappropriate in the circumstances of this case, I dissent from the majority opinion insofar as it holds that Ms. Bayless's claim was timely.

The majority opinion states that it is not "willing to engage in speculation" that the government could successfully contest its decision. Majority Op. at 1247 n. 9. It assumes that there is nothing that the government could have presented or argued that could change the result. But not only does such an assumption run contrary to our tradition of providing notice and an opportunity to respond before rendering a ruling, it suffers from the sin of speculation attributed to my view. I would rather speculate in favor of affording due process than speculate to support its denial.

So let me suggest how the government might have grounds to oppose a motion for partial summary judgment by Ms. Bayless. For one thing, the majority opinion assumes the accuracy of all of Ms. Bayless's testimony. How can we be certain that the government does not have evidence that would challenge that accuracy? There would have been no reason for the government to put on such evidence in support of its motion for summary judgment. After all, it needed to base its motion on undisputed facts. Impeaching Ms. Bayless's testimony would have accomplished nothing, because the court would nevertheless have to credit her veracity and the reliability of her memory. Accordingly, the government set forth a summary of her deposition testimony as stating undisputed facts. Those undisputed facts, however, were undisputed "only for the purpose of presenting a legal theory, which, if accepted by the court, would entitle the government to a judgment." United States v. Mills, 372 F.2d 693, 697 (10th Cir.1966); see 10A Charles A. Wright et al., Federal Practice and Procedure § 2720 (3d ed. 1998) ("[A] party moving for summary judgment concedes the absence of a factual issue and the truth of the nonmoving party's allegations only for purposes of his own motion." (emphasis added)). On the other hand, if Ms. Bayless had moved for partial summary judgment, perhaps the government could have defeated the motion by presenting evidence contrary to hers on the facts essential to her motion.

The approach of the majority opinion will work a sea change in summary-judgment practice. A party seeking summary judgment will now have to protect itself by presenting all evidence that could impeach evidence favoring the opposing party. Such evidence, of course, is irrelevant to the summary-judgment motion because the court reviewing the motion must view the evidence in the light most favorable to the nonmovant. But if the moving party does not include impeaching evidence, it could face a court that follows the precedent established by the majority opinion and (1) construes the nonmovant's opposition to the summary-judgment motion as a request for a judicial determination that the nonmovant is entitled to summary judgment (or at least partial summary judgment on the issue) and (2) refuses "to engage in speculation that the [movant] might have additional cards in its hand ... which might turn the tide in its favor at trial." Majority Op. at 1247 n. 9.

The only advantage I can see to such an unfair procedure is that it will save paper otherwise used in litigating cross-motions for summary judgment. Why file a cross-motion when you have a better chance of a favorable ruling by just opposing the other party's motion? If Ms. Bayless had moved for a partial summary judgment that her complaint was timely, the government would then have been alerted to the need to put on any evidence it had that would cast doubt on Ms. Bayless's assertions relevant to the limitations issue. By not so moving, Ms. Bayless would avoid alerting the government that it needed to respond. How much better simply to oppose the government's motion and not take a chance that her account could be impeached? If there is any precedent for the approach of the majority opinion, I am not aware of it.

Moreover, even on the limited evidence now before us, the government could marshal arguments that Ms. Bayless's rendition of facts (which had to be accepted as true for purposes of the government's motion) is actually unlikely and that a reasonably diligent person in Ms. Bayless's situation would have discovered the cause of her ailments significantly sooner than she did. It is important to keep in mind the legal standard. The discovery-rule analysis depends on what the plaintiff knew as time progressed. "The exercise of reasonable diligence is an ongoing process. What is required at any particular time depends on what one has notice of at that time. When additional information is obtained, the standard of reasonable diligence may demand action that would not have been demanded without receipt of the information." United States v. Denny, 694 F.3d 1185, 1190 (10th Cir.2012). One could argue from the record evidence that in early 2005 Ms. Bayless had acquired information mandating that she obtain scientific testing of whether she had been exposed to sarin gas but that she had delayed an unreasonably long time before doing so.

If we examine the evidence at hand in the light most favorable to the United States rather than, as in the majority opinion, the light most favorable to Ms. Bayless, we get the following picture. By early 2005 Ms. Bayless had learned (1) that where she had been working shortly before her ailments first arose was near a military facility where sarin gas was present and (2) that her ailments were similar to those of a man who had been exposed to sarin at that facility. On May 4, 2005, she wrote to Mr. Queen, a clinical nutritionist who had been treating her, that she was "pretty sure" that her problems had been caused by exposure to nerve agents, probably sarin gas. App. to Aplt. Br. at 116. I do not think that the majority would disagree that a reasonably diligent person would promptly follow up on that belief. So what did she do?

On February 1, 2005, Mr. Queen had told her that the Institute of Orthomolecular Medicine "would probably have a way to test for [nerve agents]" and that "the information may be very useful." She admits, however, that she did not follow up on this advice.

Also, in July, August, or September of 2005, Mr. Queen had her take huperzine, saying that it would relieve her symptoms if they were caused by sarin. After a few days, however, her symptoms only got worse, so she quit using it. If that were the only reason that Ms. Bayless quit investigating whether sarin had caused her problems, I would not agree that she was acting as a reasonable person in that regard. Mr. Queen was not a doctor. And huperzine is a Spanish club moss, not a prescription medication. A reasonable person would not have relied on the huperzine trial as reason not to pursue sarin as a cause.

The majority opinion suggests that a negative PCR test in December 2006 also deterred Ms. Bayless from investigating sarin gas as the cause of her condition. But the PCR test was for exposure to biological weapons. Sarin is not a biological weapon.

What about the acetylcholine test? Ms. Bayless testified that she had learned from Internet research by February 2005 that such a test could help decide whether she was suffering from exposure to nerve gas. And a test of that nature (the record before us leaves unclear whether it was an identical test) is ultimately what she bases her present claim on. To act with reasonable diligence she would have needed to pursue scientific testing for exposure to nerve gas in early 2005. Contrary to the majority opinion, however, there is a reasonable question whether she did. Ms. Bayless testified that because of the importance of the test, she had it performed in 2005. But her recollection may have been mistaken. At her deposition she, quite understandably, expressed uncertainty about a number of matters in her medical history, including events in 2005. Turning specifically to the acetylcholine test, the record before us contains no report of such a test in 2005 or any mention of such a test in the medical records for that year. (Of course, the parties may have documentation of such a test. But, in light of Ms. Bayless's testimony, it would have been irrelevant to any issue on appeal, so there would have been no reason to include it in the appendix. Guessing about what evidence there may be is a hazard in addressing an issue never addressed by the parties on appeal.) Although Ms. Bayless testified that the test was performed by Dr. Moody, his office notes clearly show that her last visit with him was on May 3, 2005, the day before she wrote the message to Mr. Queen that she was "pretty sure" that sarin was the culprit. Dr. Moody's notes for May 3 do include the notation "test results," but we do not know what the tests were. Id. at 113. If Dr. Moody gave her the (negative) result of the acetylcholine test on that date, it would be strange for her to write her message to Mr. Queen the next day. And if the notation was to indicate that Dr. Moody should call her with the results, it would be strange for her to write her "pretty sure" message without awaiting the results or even mentioning that she was awaiting results. In addition, Ms. Bayless testified that sometime between July and September of 2005 she tried huperzine for a few days. But why was she still being tested for sarin if she had earlier received test results (the acetylcholine test) that convinced her that sarin was not the cause of her ailments?

Thus, a reasonable factfinder could find that Ms. Bayless's decision in 2005 not to pursue sarin was the result of the huperzine trial, not a nonexistent acetylcholine test. And if that were the case, I would say that Ms. Bayless had not acted with reasonable diligence and that a reasonably diligent person in her situation would have discovered the sarin connection substantially before she actually did.

For these reasons, I dissent from the decision of the majority that as a matter of law Ms. Bayless's claim was timely.