*1083 Relator Billy Joe Hunt filed a qui tam action alleging that his employer The Parsons Corporation and another entity, Cochise Consultancy, Inc., violated the False Claims Act ("FCA"), 31 U.S.C. §§ 3729 - 33, by submitting to the United States false or fraudulent claims for payment. Hunt filed his action more than six years after the alleged fraud occurred but within three years of when he disclosed the fraud to the government. In this appeal, we are called upon to decide whether Hunt's FCA claim is time barred. To answer this question, we must construe the FCA's statutory provision that requires a civil action alleging an FCA violation to be brought within the later of:

• "6 years after the date on which the violation ... is committed," 31 U.S.C. § 3731 (b)(1), or

• "3 years after the date when facts material to the right of action are known or reasonably should have been known by the official of the United States charged with responsibility to act in the circumstances, but in no event more than 10 years after the date on which the violation is committed," id. § 3731(b)(2).

The question we answer today, which is one of first impression, is whether § 3731(b)(2) 's three year limitations period applies to a relator's FCA claim when the United States declines to intervene in the qui tam action.

The district court concluded that the limitations period in § 3731(b)(2) is inapplicable in such cases and thus Hunt's claim is time barred. After careful consideration of the statutory scheme, we hold that § 3731(b)(2) 's three year limitations period applies to an FCA claim brought by a relator even when the United States declines to intervene. Further, because the FCA provides that this period begins to run when the relevant federal government official learns of the facts giving rise to the claim, when the relator learned of the fraud is immaterial for statute of limitations purposes. Here, it is not apparent from the face of Hunt's complaint that his claim is untimely because his allegations show that he filed suit within three years of the date when he disclosed facts material to the right of action to United States officials and within ten years of when the fraud occurred. The district court therefore erred in dismissing his complaint. We reverse and remand to the district court for further proceedings.

I. FACTUAL BACKGROUND

A. The Fraudulent Scheme

Hunt alleges that Parsons and Cochise (the "contractors") defrauded the United States Department of Defense for work they performed as defense contractors in Iraq. 1 The Department of Defense awarded Parsons a $60 million contract to clean up excess munitions in Iraq left behind by retreating or defeated enemy forces. Hunt worked for Parsons in Iraq *1084 on the munitions clearing contract, managing the project's day-to-day operations. One facet of the contract required Parsons to provide adequate security to its employees, its subcontractors, and others who were working on the munitions clearing project. Parsons relied on a subcontractor to provide the security services.

After seeking bids for the security subcontract, a Parsons committee awarded it to ArmorGroup. But an Army Corps of Engineers contracting officer in Iraq whom Cochise had bribed with trips and gifts, Wayne Shaw, was determined to override this decision and have the subcontract awarded to Cochise. Shaw directed Hunt to have Hoyt Runnels, another Parsons employee who served on the committee that selected ArmorGroup, issue a directive awarding Cochise the subcontract. When Hunt did so, Runnels refused to issue the directive, explaining that such a directive had to come from the Corps.

Shaw then created a forged directive rescinding the award to ArmorGroup and awarding the subcontract to Cochise. The directive had to be signed by Steven Hamilton, another Corps contracting officer. Hamilton, who was legally blind, relied on Shaw to describe the document he was signing. Shaw did not disclose that the directive rescinded the award to ArmorGroup so that the subcontract could be awarded to Cochise.

After Hamilton signed the directive, Shaw directed Runnels to execute it. Runnels again refused because he believed the award to Cochise had been made in violation of government regulations. Shaw threated to have Runnels fired. Two days later, Hamilton learned that the directive Shaw had him sign rescinded the award to ArmorGroup and awarded Cochise the subcontract. Hamilton immediately rescinded his directive awarding the subcontract to Cochise.

After Runnels refused to follow Shaw's directive to award the subcontract to Cochise, another Parsons employee, Dwight Hill, replaced Runnels and was given responsibility for awarding the security subcontract. Rather than give the subcontract to ArmorGroup, Hill awarded it to Cochise through a no-bid process. Hill justified using a no-bid process by claiming there was an urgent and immediate need for convoy services and then defended the choice of Cochise to fill this immediate need by asserting that Cochise had experience that other security providers lacked. But Hunt alleges that Hill selected Cochise because he was its partner in the fraudulent scheme.

From February through September 2006, Cochise provided security services under the subcontract. Each month the United States government paid Cochise at least $1 million more than it would have paid ArmorGroup had ArmorGroup been awarded the subcontract. The government incurred other additional expenses as well. For example, armored vehicles were needed to provide the security services, and because Cochise had no such vehicles, the government paid more than $2.9 million to secure the vehicles. In contrast, ArmorGroup would have supplied its own armored vehicles, saving the government millions of dollars. In September 2006, when Shaw rotated out of Iraq, Parsons immediately reopened the subcontract for bidding and awarded it to ArmorGroup.

Several years later, Hunt reported the fraud to the United States government. On November 30, 2010, FBI agents interviewed Hunt about his role in a separate kickback scheme. During the interview, Hunt told the agents about the contractors' fraudulent scheme involving the subcontract for security services. For his role in the separate kickback scheme, Hunt was charged with federal crimes, pled *1085 guilty, and served ten months in federal prison.

B. Procedural History

After his release from prison, on November 27, 2013, Hunt filed under seal in federal district court an FCA complaint against the contractors. Hunt set forth two theories why the claims the contractors submitted for payment qualified as false claims under the FCA. First, he alleged that Cochise fraudulently induced the government to enter into the subcontract to purchase Cochise's services by providing illegal gifts to Shaw and his team. He alleged that Parsons, through Hill, conspired with Cochise and Shaw to rig the bidding process for the subcontract. Second, Hunt alleged that the contractors had a legal obligation to disclose credible evidence of improper conflicts of interest and payment of illegal gratuities to the United States but failed to do so.

After the United States declined to intervene, Hunt's complaint was unsealed. The contractors moved to dismiss, arguing that the claim was time barred under the six year limitations period in 31 U.S.C. § 3731 (b)(1), and Hunt had waited more than seven years after the fraud occurred to file suit. Hunt responded that his claim was timely under the limitations period in § 3731(b)(2) because he had filed suit within three years of when the government learned of the fraud at his FBI interview and ten years of when the fraud occurred. The district court disagreed, concluding that § 3731(b)(2) 's limitations period was either (1) unavailable to Hunt because the United States had declined to intervene or (2) expired because it began to run when Hunt learned of the fraud. The district court then granted the motions to dismiss, finding Hunt's claim untimely under § 3731(b)(1) 's limitation period because it was apparent from the face of Hunt's complaint that he failed to file suit within six years of when the fraud occurred. This is Hunt's appeal.

II. STANDARD OF REVIEW

We review de novo a district court's dismissal of a complaint for failure to state a claim upon which relief can be granted. Am. Dental Ass'n v. Cigna Corp. , 605 F.3d 1283 , 1288 (11th Cir. 2010). A dismissal for failure to state a claim on statute of limitations grounds is appropriate "only if it is apparent from the face of the complaint that the claim is time-barred." La Grasta v. First Union Sec., Inc. , 358 F.3d 840 , 845 (11th Cir. 2004) (internal quotation marks omitted). "We review the district court's interpretation and application of statutes of limitations de novo ." Ctr. for Biological Diversity v. Hamilton , 453 F.3d 1331 , 1334 (11th Cir. 2006).

III. BACKGROUND ON THE FCA

Before addressing whether Hunt's claim is timely, we pause to provide some necessary background information about the roles of the government and the private plaintiff in a qui tam suit and to discuss the relevant FCA provisions. The FCA was enacted in 1863 to "stop[ ] the massive frauds perpetrated by large contractors during the Civil War." Universal Health Servs., Inc. v. United States ex rel. Escobar , --- U.S. ----, 136 S.Ct. 1989 , 1996, 195 L.Ed.2d 348 (2016) (internal quotation marks omitted). These contractors billed the United States "for nonexistent or worthless goods, charged exorbitant prices for goods delivered, and generally robbed in purchasing the necessities of war." Id. (internal quotation marks omitted). In response, Congress passed the original FCA, which imposed civil and criminal liability for fraud on the government, subjecting *1086 violators to double damages, forfeiture, and imprisonment. Id.

Since 1863, Congress repeatedly has amended the FCA. Today, the FCA continues to prohibit making false claims for payment to the United States. See 31 U.S.C. § 3729 (a). But unlike the original FCA that provided for both civil and criminal liability, violators today face only civil liability, which subjects them to treble damages and civil penalties. 2 Id.

Section 3730 of the FCA sets forth three different enforcement mechanisms for a violation of the Act. Section 3730(a) provides that the Attorney General may sue a violator in a civil lawsuit. Section 3730(b) allows a private plaintiff, known as a relator, to bring a qui tam action in the name of the United States against a violator. Section 3730(h) creates a private right of action for an individual whose employer retaliated against him for assisting an FCA investigation or proceeding.

This appeal concerns the second mechanism, a qui tam action brought by a relator under § 3730(b). In a qui tam action, the relator "pursues the government's claim against the defendant, and asserts the injury in fact suffered by the government." Stalley ex rel. United States v. Orlando Reg'l Healthcare Sys., Inc. , 524 F.3d 1229 , 1233 (11th Cir. 2008). 3 In bringing a qui tam action, the relator "in effect, su[es] as a partial assignee of the United States." Vt. Agency of Nat. Res. v. United States ex rel. Stevens , 529 U.S. 765 , 773 n.4, 120 S.Ct. 1858 , 146 L.Ed.2d 836 (2000) (emphasis omitted).

Special procedures apply when a relator brings an FCA action; these procedures afford the government the opportunity to intervene and assume primary control over the litigation. A relator who initiates an FCA action must file her complaint under seal and serve it only on the United States. 31 U.S.C. § 3730 (b)(2). While the lawsuit remains under seal, the United States has the opportunity to investigate and decide whether to intervene as a party. 4 Id. During this period, the United States may serve a civil investigative demand upon any person believed to be in possession of documents or information relevant to an investigation of false claims, requiring that person to produce documents, answer interrogatories, or give oral testimony. Id. § 3733(a)(1). In addition, the United States may meet with the relator and her attorney, giving the government an opportunity to ask questions to assess the strengths and weaknesses of the case and the relator a chance to assist the government's investigation. 5

*1087 If the United States decides to intervene, the government acquires "primary responsibility for prosecuting the action," although the relator remains a party. Id. § 3730(c)(1). In contrast, if the United States declines to intervene, the relator may proceed with the action alone on behalf of the government, but the United States is not a party to the action. Id. § 3730(c)(3).

Although the United States is not a party to a non-intervened case, it nevertheless retains a significant role in the litigation. The government may request to be served with copies of all pleadings and deposition transcripts, seek to stay discovery if it "would interfere with the Government's investigation or prosecution of a criminal or civil matter arising out of the same facts," and veto a relator's decision to voluntarily dismiss the action. Id. § 3730(b)(1), (c)(3), (c)(4). Additionally, the court may permit the government to intervene later "upon a showing of good cause." Id. § 3730(c)(3).

Any recovery obtained from a defendant in an FCA qui tam action belongs to the United States, regardless of whether the government has intervened. The relator is entitled to a portion of the recovery, however. Id. § 3730(d). Because the relator receives a share of the government's proceeds, he "is essentially a self-appointed private attorney general, and his recovery is analogous to a lawyer's contingent fee." United States ex rel. Milam v. Univ. of Tex. M.D. Anderson Cancer Ctr. , 961 F.2d 46 , 49 (4th Cir. 1992) ; see Cook Cty. v. United States ex rel. Chandler , 538 U.S. 119 , 122, 123 S.Ct. 1239 , 155 L.Ed.2d 247 (2003) (explaining that a relator sues in the name of the government "with the hope of sharing in any recovery"). By allowing a relator to bring a qui tam action and share in the government's recovery, the FCA creates an economic incentive to encourage "citizens to come forward with knowledge of frauds against the government." Milam , 961 F.2d at 49 .

The size of the relator's share depends upon whether the United States intervenes. In an intervened case, the relator usually is entitled to between 15 and 25 percent of the proceeds, as well as reasonable expenses, attorney's fees, and costs. 31 U.S.C. § 3730 (d)(1). In a non-intervened case, the relator's share usually is greater: between 25 and 30 percent of the proceeds, as well as reasonable expenses, attorney's fees, and costs. Id. § 3730(d)(2).

Even though the relator receives a smaller share in an intervened case, relators generally try to persuade the United States to intervene because the government's intervention makes it far more likely that there will be a recovery. When the United States elects to intervene, about 90 percent of the time the case generates a recovery, either through settlement or a final judgment. But only about 10 percent of non-intervened cases result in recovery. 6

*1088 See David Freeman Engstrom, Public Regulation of Private Enforcement: Empirical Analysis of DOJ Oversight of Qui Tam Litigation Under the False Claims Act , 107 Nw. U. L. Rev. 1689 , 1720-21 (2013). Indeed, when the government declines to intervene, more than 50 percent of the time the relator decides not to proceed and voluntarily dismisses the action. See id. at 1717-18 .

IV. ANALYSIS

With this general background in mind, we now turn to the issue in this case: whether it is apparent from the face of Hunt's complaint that his FCA claim is time barred. To answer this question, we must interpret the FCA's statute of limitations provision, which creates two limitations periods that potentially apply:

(b) A civil action under section 3730 may not be brought-

(1) more than 6 years after the date on which the violation of section 3729 is committed, or

(2) more than 3 years after the date when facts material to the right of action are known or reasonably should have been known by the official of the United States charged with responsibility to act in the circumstances, but in no event more than 10 years after the date on which the violation is committed,

whichever occurs last.

31 U.S.C. § 3731 (b). Because it is apparent from the face of Hunt's complaint that he failed to file his action within the six year limitations period of § 3731(b)(1), this case turns on whether Hunt can avail himself of § 3731(b)(2). To determine whether § 3731(b)(2) applies, we must address whether its limitations period is available when the United States declines to intervene and, if so, whether the limitations period is triggered when the relator knew or should have known facts material to his claim.

A. Section 3731(b)(2) Applies When the United States Declines to Intervene.

The primary question before us is whether Congress intended to allow relators in non-intervened cases to rely on § 3731(b)(2) 's limitations period. We must begin "where courts should always begin the process of legislative interpretation, and where they often should end it as well, which is with the words of the statutory provision." Harris v. Garner , 216 F.3d 970 , 972 (11th Cir. 2000) (en banc). In considering the text, we bear in mind that "[a] provision that may seem ambiguous in isolation is often clarified by the remainder of the statutory scheme." Koons Buick Pontiac GMC, Inc. v. Nigh , 543 U.S. 50 , 60, 125 S.Ct. 460 , 160 L.Ed.2d 389 (2004) (internal quotation marks omitted). We look to "the whole statutory text, considering the purpose and context of the statute, and consulting any precedents or authorities *1089 that inform the analysis." Dolan v. U.S. Postal Serv. , 546 U.S. 481 , 486, 126 S.Ct. 1252 , 163 L.Ed.2d 1079 (2006). As part of this inquiry, we also consider the canons of statutory construction. CBS Inc. v. PrimeTime 24 Joint Venture , 245 F.3d 1217 , 1225 (11th Cir. 2001). Legislative history may prove helpful when the statutory language remains ambiguous after considering "the language itself, the specific context in which that language is used, and the broader context of the statute as a whole." Robinson v. Shell Oil Co. , 519 U.S. 337 , 341, 117 S.Ct. 843 , 136 L.Ed.2d 808 (1997).

We conclude that the phrase "civil action under section 3730" in § 3731(b) refers to civil actions brought under § 3730 that have as an element a violation of § 3729, which includes § 3730(b) qui tam actions when the government declines to intervene. Section § 3731(b) begins by providing that its limitations periods apply to "[a] civil action under section 3730." 31 U.S.C. § 3731 (b). A non-intervened cases is a type of civil action under § 3730. See id. § 3730(b)(1) (permitting any person to bring a civil action alleging a violation of § 3729 ); id. § 3730(c)(3) (allowing a relator to continue to conduct a qui tam action after the United States declines to intervene). And nothing in § 3731(b)(2) says that its limitations period is unavailable to relators when the government declines to intervene. In the absence of such language, we conclude that the text supports allowing relators in non-intervened cases to rely on § 3731(b)(2) 's limitations period.

To ascertain its meaning, we must, of course, view § 3731(b)(2) in the broader statutory context. Looking to the statutory context, the Supreme Court has recognized that the phrase "[a] civil action under section 3730" did not refer to all types of § 3730 civil actions because it excluded retaliation actions brought under § 3730(h). Graham Cty. Soil & Water Conservation Dist. v. United States ex rel. Wilson , 545 U.S. 409 , 415, 125 S.Ct. 2444 , 162 L.Ed.2d 390 (2005). 7 In Graham County , the Supreme Court considered whether § 3731(b)(1) 's six year limitations period-which begins to run when the defendant submits a false claim-applied to an employee's § 3730(h) retaliation claim alleging that her employer forced her to resign after she assisted federal officials investigating her employer for submitting false claims to the United States. Id. at 413-14 , 125 S.Ct. 2444 . On its face, § 3731(b) appeared to apply to § 3730(h) retaliation actions, which were a type of civil action under § 3730. Id. at 415 , 125 S.Ct. 2444 . Relying on statutory context, the Court nonetheless concluded that § 3731(b) 's literal text was ambiguous as to whether the phrase "[a] civil action under section 3730" included § 3730(h) retaliation actions. Id. at 417 , 125 S.Ct. 2444 . The Court observed that § 3731(b)(1) 's limitations period was triggered by the defendant's submission of a false claim. Id. at 415 , 125 S.Ct. 2444 . But a plaintiff bringing a retaliation claim under § 3730(h) did not need to allege or prove that the defendant actually submitted a false claim because an employer can be liable for retaliating against an employee *1090 who assists with an investigation or civil action even if the employer is innocent. Id. at 416 , 125 S.Ct. 2444 . This tension in applying § 3731(b)(1) 's limitation period to retaliation actions led the Court to find the statute ambiguous as to whether "action under section 3730" referred to "all actions under § 3730, or only §§ 3730(a) and (b) actions." Id.

The Supreme Court resolved this ambiguity by concluding that § 3731(b)(1) 's limitations period did not apply to retaliation claims under § 3730(h). The Court recognized that Congress generally drafted statutes of limitations to begin to run when a cause of action accrues. Id. at 418 , 125 S.Ct. 2444 . Applying § 3731(b)(1) 's limitations period to an FCA retaliation action would violate this general rule because the limitations period would begin to run when the employer committed the actual or suspected FCA violation, not when it retaliated against the employee. This interpretation could lead to the odd result that a plaintiff's retaliation claim was time barred before the employer took any retaliatory action. Id. at 420-21 , 125 S.Ct. 2444 . To "avoid[ ] these counterintuitive results," the Court construed "civil action under section 3730" to "mean[ ] only those civil actions under § 3730 that have as an element a violation of section 3729, that is, §§ 3730(a) and (b) actions." Id. at 421-22 , 125 S.Ct. 2444 (internal quotation marks omitted). 8 Graham County thus made clear that to determine whether § 3731(b)(2) includes qui tam actions where the United States declines to intervene, we must consider the text of § 3731(b)(2) in the relevant statutory context. But nothing in Graham County directly addressed whether the statutory context shows that § 3731(b)(2) 's limitations period is available only when the government is a party.

Here, the contractors raise several arguments contending that the statutory context and the canons of statutory construction show that Congress intended for § 3731(b)(2) to be unavailable to relators in non-intervened cases. They claim that allowing a relator in a non-intervened action to rely on a limitations period that is triggered by a government official's knowledge would lead to absurd results and render a portion of § 3731(b) superfluous. We reject each of these arguments. The text of § 3731(b)(2), when viewed in context, shows that § 3731(b)(2) is available to relators when the government declines to intervene. But even if we were to conclude that § 3731(b)(2) is ambiguous making it *1091 appropriate to consider legislative history, as the contractors urge us to do, we still would conclude that § 3731(b)(2) is available to relators when the government declines to intervene.

1. We Reject that Allowing a Relator in a Non-Intervened Case to Rely on § 3731(b)(2) 's Limitations Period Is Absurd.

The contractors' primary argument is that the statutory context shows that § 3731(b)(2) is available only when the United States is a party to the case because the limitations period is triggered by a federal official's knowledge. They argue that Congress must have intended such a limitations period to be available only when the government is a party to the case because to apply a limitations period triggered by a federal official's knowledge when the United States is not a party would create a "bizarre scenario." Parsons' Br. at 12 (quoting United States ex rel. Sanders v. N. Am. Bus Indus., Inc. , 546 F.3d 288 , 293 (4th Cir. 2008) ). Put differently, they argue that reading § 3731(b)(2) to apply to non-intervened actions would lead to an absurd result. Of course, we should refrain from interpreting a statute in a way that "produces a result that is not just unwise but is clearly absurd." CBS , 245 F.3d at 1228 (internal quotation marks omitted). But we have cautioned that the absurdity doctrine is "rarely applied" to avoid having "clearly expressed legislative decisions ... be subject to the policy predilections of judges." Id. (internal quotation marks omitted).

This case presents no such rare instance when the absurdity doctrine applies. Certainly, it is generally the case that a discovery-based limitations period begins to run when a party -the plaintiff-knew or should have known about the fraud or claim. See, e.g., Merck & Co. v. Reynolds , 559 U.S. 633 , 637, 130 S.Ct. 1784 , 176 L.Ed.2d 582 (2010) (recognizing that a securities fraud claim accrued when the plaintiff knew or should have known the facts constituting the violation); see also Restatement (Second) of Torts § 899(e) (statute of limitations begins to run when "the injured person has knowledge or reason to know of the facts"). We cannot say that in the unique context of an FCA qui tam action, 9 however, it would be absurd to peg a limitations period to a federal official's knowledge unless the United States brings the action or chooses to intervene. We reject the contractors' absurdity argument because even though the United States is not a party to a non-intervened qui tam action, the United States remains the real party in interest and retains significant control over the case.

Even in a non-intervened case, the relator brings the suit as the partial assignee of the United States and asserts a claim based on injury suffered by the United States as the victim of the fraud. United States ex rel. Eisenstein v. City ofNew York , 556 U.S. 928 , 934-35, 129 S.Ct. 2230 , 173 L.Ed.2d 1255 (2009). Importantly, as the victim of the fraud, the United States-not the relator-is entitled to the bulk of the recovery. See 31 U.S.C. § 3730 (d)(2). Given the government's primary interest in a non-intervened qui tam action, Congress carved out for it a formal role, allowing it to intervene at any time upon a showing of good cause, request service of pleadings and deposition transcripts, seek to stay discovery if it "would interfere with the Government's investigation or prosecution of a criminal or civil matter arising out of the same facts," and *1092 veto a relator's decision to voluntarily dismiss the action. Id. § 3730(b)(1), (c)(3), (c)(4). Given this unique role, we cannot say that it would be absurd for Congress to peg the start of the limitations period to the knowledge of a government official even when the United States declines to intervene.

The contractors argue that allowing a relator in a non-intervened case to rely on § 3731(b)(2) 's limitations period conflicts with the Supreme Court's decision in Eisenstein . In Eisenstein , the relators in a non-intervened case filed a notice of appeal 54 days after the district court entered a final judgment dismissing their claims. 556 U.S. at 930 , 129 S.Ct. 2230 . Although parties normally have 30 days to file a notice of appeal, the relators argued that they could avail themselves of the 60 day deadline that applies when the United States is a party to the action. Id. at 930-31 , 129 S.Ct. 2230 . The Supreme Court rejected this argument and affirmed the dismissal of the appeal, holding that the United States is not a party to a qui tam action when it declines to intervene. Id. at 937 , 129 S.Ct. 2230 . But our decision today in no way relies on the United States being a party to the non-intervened case, and nothing in Eisenstein addressed whether the United States' non-party status means that the limitations period in § 3731(b)(2) is unavailable to relators in non-intervened cases.

We recognize that our decision to reject the absurdity doctrine is at odds with the published decisions of two other circuits. See Sanders , 546 F.3d at 293 ("Congress intended Section 3731(b)(2) to extend the FCA's default six-year period only in cases in which the government is a party, rather than to produce the bizarre scenario in which the limitations period in a relator's action depends on the knowledge of a nonparty to the action."); United States ex rel. Sikkenga v. Regence Bluecross Blueshield of Utah , 472 F.3d 702 , 726 (10th Cir. 2006) ("Surely, Congress could not have intended to base a statute of limitations on the knowledge of a non-party.").

These cases do not persuade us. They reflexively applied the general rule that a limitations period is triggered by the knowledge of a party. They failed to consider the unique role that the United States plays even in a non-intervened qui tam case. In light of this role, we cannot say that it would be absurd or "bizarre" to peg the limitations period to the knowledge of a government official when the government declines to intervene. We disagree that Congress, by specifying that § 3731(b)(2) 's limitations period is triggered by the knowledge of a United States official, necessarily intended that this limitations period be available only in § 3730 civil actions where the United States is a party and not in non-intervened qui tam actions. 10 We thus cannot say that the statutory *1093 context shows that § 3731(b)(2) 's limitations period is unavailable to relators in non-intervened qui tam actions.

2. Our Interpretation Does Not Render a Portion of § 3731(b) Superfluous.

The contractors, relying on a canon of construction, next argue that to give meaning to the entirety of § 3731(b), we must construe § 3731(b)(2) to exclude non-intervened cases. Certainly, "a statute ought, upon the whole, to be so construed that, if it can be prevented, no clause, sentence, or word shall be superfluous, void, or insignificant." TRW Inc. v. Andrews , 534 U.S. 19 , 31, 122 S.Ct. 441 , 151 L.Ed.2d 339 (2001) (internal quotation marks omitted). But this canon does not apply when a statutory provision would remain operative under the interpretation in question in at least some situations. See Black Warrior Riverkeeper, Inc. v. Black Warrior Minerals, Inc. , 734 F.3d 1297 , 1304 (11th Cir. 2013).

The contractors assert that if relators have three years from the date when the government learned of the fraud to file suit under § 3731(b)(2), relators will always delay telling the government about the fraud to increase the damages in the case. Therefore, they say, the limitations period in § 3731(b)(1), which expires six years after the date when the violation occurred, will never apply, rendering the provision meaningless. We disagree. The contractors overlook that other provisions of the FCA create strong incentives to ensure that relators promptly report fraud.

A relator who waits to report a fraud risks recovering nothing or having his relator's share decreased. The relator's claim may be barred if another relator beats him to the courthouse with an FCA claim based on the same facts, 31 U.S.C. § 3730 (b)(5), or if the allegations or transactions are publicly disclosed either in a federal hearing where the government was a party or in a news report, unless the relator was the original source of the information, id. § 3730(e)(4). And because § 3731(b)(2) 's limitations period begins to run when the relevant government officials learns about the fraud from any source, a relator who delays reporting the fraud to the government also runs the risk that the government will learn about the fraud from another source and thus that § 3731(b)(2) 's three year period will expire before the relator files suit. But even if there were no risk that the government could learn of the fraud from another source, a relator still would have an incentive to report fraud promptly because the court in setting the relator's share may consider whether he "substantially delayed in reporting the fraud or filing the complaint."

*1094 United States ex rel. Shea v. Verizon Commc'ns, Inc. , 844 F.Supp.2d 78 , 89 (D.D.C. 2012).

Looking at the FCA as a whole, we conclude that relators who can rely on the limitations period in § 3731(b)(2) will still have sufficient incentive to report fraud promptly. Because relators will continue to report fraud promptly and under § 3731(b)(2) suit must be filed within three years of the fraud being reported, there will be cases in which § 3731(b)(1) 's six year limitations period will expire later. We thus reject the contractors' argument that our reading of the FCA would render superfluous one of its provisions.

3. To the Extent that Legislative History is Relevant, It Bolsters Our Conclusion.

The contractors argue that the legislative history shows that § 3731(b)(2) 's limitations period is unavailable to a relator when the United States declines to intervene. Assuming that the statutory language, after viewing it in light of the statutory context and the canons of construction, remains ambiguous such that a resort to legislative history is appropriate, see United States v. Alabama , 778 F.3d 926 , 939 (11th Cir. 2015), we cannot agree that the relevant Congressional records undermine our interpretation of § 3731(b)(2).

Congress added the limitations period in § 3731(b)(2) to the FCA in 1986. False Claims Amendments Act of 1986 ("1986 FCA Amendments"), Pub. L. No. 99-562, 100 Stat. 3153 (1986). The legislative history reveals that one of the broad purposes of the 1986 FCA Amendments was to "encourage more private enforcement suits." S. Rep. No. 99-345 at 23-24 (1986). This purpose is consistent with Congress's historical use of qui tam rights of action to create incentives for private individuals to help root out fraud against the government. See United States ex rel. Williams v. NEC Corp. , 931 F.2d 1493 , 1497 (11th Cir. 1991). Allowing relators to continue to pursue FCA claims even after the government declines to intervene is consistent with the broad underlying purpose of the FCA because it creates the potential for "more fraud [to] be discovered, more litigation [to] be maintained, and more funds [to] flow back into the Treasury." Milam , 961 F.2d at 49 .

The contractors argue that we should not infer Congressional intent to extend the limitations period for non-intervened cases because in the legislative history for the 1986 FCA Amendments Congress indicated that qui tam actions must be brought shortly after the fraud occurred. To support their position, the contractors point to the following portion of the Senate Committee Report, which quotes from the reasoning in a Supreme Court decision:

[The FCA] is intended to protect the Treasury against the hungry and unscrupulous host that encompasses it on every side, and should be construed accordingly. It was passed upon the theory, based on experience as old as modern civilization, that one of the least expensive and most effective means of preventing frauds on the Treasury is to make the perpetrators of them liable to actions by private persons acting, if you please, under the strong stimulus of personal ill will or the hope of gain. Prosecutions conducted by such means compare with the ordinary methods as the enterprising privateer does to the slow-going public vessel.

S. Rep. No. 99-345, at 11 (quoting United States ex rel. Marcus v. Hess , 317 U.S. 537 , 541 n.5, 63 S.Ct. 379 , 87 L.Ed. 443 (1943) ).

The contractors argue this language shows that Congress allowed relators to bring qui tam actions under the FCA because *1095 relators are able to expose fraud more rapidly than the United States can discover it, from which they infer that Congress intended for a shorter limitations period to apply when the United States was not a party to the case. But nothing in this statement addresses the length of time that a relator should have to bring a qui tam action or whether the limitations period should depend on the government's decision to intervene. And so we fail to see how this legislative history supports the contractors' position that a shorter limitations period should apply when the government declines to intervene.

All told, there is little legislative history for § 3731(b)(2). And the few references there are do not directly address the question before us. The contractors point to a floor statement from Senator Charles Grassley and testimony from Assistant Attorney General Richard K. Willard before a House subcommittee. But neither piece of legislative history is particularly helpful.

Senator Grassley said in a floor statement that Congress borrowed the language in § 3731(b)(2) from 28 U.S.C. § 2416 , which sets forth the limitations period that generally applies to other actions brought by the United States. See 132 Cong. Rec. 20,536 (1986) (statement of Sen. Grassley). Senator Grassley's statement reflects that Congress borrowed the language "facts material to the right of action are known or reasonably should have been known by the official of the United States charged with responsibility to act" from 28 U.S.C. § 2416 . See 28 U.S.C. § 2416 (c) ; 31 U.S.C. § 3731 (b)(2). But we disagree with the inference the contractors draw from this fact: that Congress intended to make the statute of limitations in § 3731(b) available only when the United States was a party.

To understand 28 U.S.C. § 2416 , we must also look to § 2415. Section 2415 establishes various limitations periods for certain categories of claims "brought by the United States or an officer or agency thereof," such as contract or tort claims. 28 U.S.C. § 2415 (a), (b). Section 2416 tolls the limitations period for the United States to bring such claims when "facts material to the right of action are not known and reasonably could not be known by an official of the United States charged with the responsibility to act in the circumstances." Id. § 2416(c). The duplicate language in § 2416 is not what specifies that a limitations period in § 2415 applies only when the United States is a party. Instead, § 2415 itself dictates that the United States must be a party for its limitations period to apply. See id. § 2415(a), (b) (stating limitations period applies only to claims "brought by the United States or an officer or agency thereof"). There is no similar language in any FCA provision expressly restricting § 3731(b)(2) 's limitations period to actions where the United States is a party. So we cannot say that by borrowing the description of the trigger for the limitations period from § 2416 Congress evinced an intent that the United States must be a party for the limitations period in § 3731(b)(2) to apply.

Turning to the committee testimony from Assistant Attorney General Willard, he explained that the purpose of § 3731(b)(2) 's limitations period was to give "us a little more flexibility in bringing some cases that otherwise would be barred." 11 The contractors construe Willard's testimony to mean that § 3731(b)(2) was intended to give the government-but *1096 not relators-more flexibility to bring FCA claims. Certainly, Willard testified that § 3731(b)(2) would extend the time period for the Attorney General to sue under the FCA. But Willard offered nothing about the intended effect of § 3731(b)(2) on qui tam actions or, more specifically, whether § 3731(b)(2) was intended to apply to qui tam actions when the government declined to intervene. Willard's testimony does not advance the ball for the contractors. See also Regan v. Wald , 468 U.S. 222 , 237, 104 S.Ct. 3026 , 82 L.Ed.2d 171 (1984) (discussing limited usefulness of testimony of witnesses to ascertain meaning of statutory language given the risk that relying on such colloquies "would open the door to the inadvertent, or perhaps even planned, undermining of the language actually voted on by Congress and signed into law by the President"). Because the legislative history does not squarely address whether Congress intended to make § 3731(b)(2) 's limitations period available to relators in non-intervened cases, we cannot agree with the contractors that the legislative history undermines our interpretation.

To wrap up, we conclude that Congress intended for § 3731(b)(2) 's limitations period to be available to relators even when the United States declines to intervene. The statutory text reflects that this limitations period applies to "[a] civil action under section 3730," and nothing in § 3731(b)(2) makes the limitations period unavailable in qui tam actions under § 3730 simply because the United States decides not to intervene. The contractors argue that because § 3731(b)(2) 's limitations period is triggered by government knowledge, Congress must have intended for it to apply only when the United States is a party to avoid absurd results. But in the unique context of a non-intervened qui tam action, we cannot say that it is absurd to apply a limitations period triggered by government knowledge. And even if the contractors are correct that we may consider legislative history, the legislative history provides no convincing support for their position.

B. The Statute of Limitations in § 3731(b)(2) Depends on the Government's Knowledge, Not the Relator's Knowledge.

Having concluded that the statute of limitations in § 3731(b)(2) is available to a relator in a non-intervened case, we must now address whether that limitations period is triggered by the knowledge of a government official or of the relator. We hold that it is the knowledge of a government official, not the relator, that triggers the limitations period.

Section 3731(b)(2) is clear that the time period begins to run when "the official of the United States charged with responsibility to act in the circumstances" knew or reasonably should have known the material facts about the fraud. 31 U.S.C. § 3731 (b)(2). Nothing in the statutory text or broader context suggests that the limitations period is triggered by the relator's knowledge. Given that the language is plain, we cannot rewrite the statute to say that the limitations period is triggered when the relator knew or should have known about the facts material to the fraud.

The Ninth Circuit nonetheless adopted such an approach, concluding that the statute of limitations is triggered by the relator's knowledge. See United States ex rel. Hyatt v. Northrop Corp. , 91 F.3d 1211 , 1217 (9th Cir. 1996). The Ninth Circuit created a new legal fiction that because the relator "sue[d] on behalf of the government," the relator became a government agent and the government official charged with responsibility to act. Id. at 1217 n.8.

*1097 Again, we find nothing in the text of § 3731(b)(2) or the statutory context to support this legal fiction. Because the text unambiguously identifies a particular official of the United States as the relevant person whose knowledge causes the limitations period to begin to run, we must reject the Ninth Circuit's interpretation as inconsistent with that text.

Applying our conclusions that § 3731(b)(2) applies in non-intervened cases and is triggered by the knowledge of a government official, not of the relator, we hold that it is not apparent from the face of Hunt's complaint that his FCA claim is untimely. Hunt alleged that the relevant government official learned the material facts on November 30, 2010 when he disclosed the fraudulent scheme to FBI agents, and he filed suit within three years of this disclosure. 12 The district court therefore erred in dismissing his complaint on statute of limitations grounds.

V. CONCLUSION

For the reasons set forth above, we reverse the district court's order dismissing Hunt's FCA claim as time barred and remand the case for further proceedings consistent with this opinion.

REVERSED AND REMANDED.