CUMMINGS, Circuit Judge.
Eunice Mathews sued the Bank of Farm-ington [Illinois] (the “Bank”) as a relator under the qui tam provision of the False Claims Act, 31 U.S.C. § 3729 et seq., in connection with an allegedly false claim the Bank made upon the former federal Farmers’ Home Administration (“FmHA”).1 Under the qui tam provisions of the False Claims Act, a private citizen can recover treble damages in a civil action brought on behalf of the government against a party that has allegedly made a false claim for payment against the United States. Id. § 3730(b). Mathews filed a qui tam lawsuit, alleging that she was the “original source” of the information about the Bank’s misrepresentation to the FmHA, and seeking damages of more than $100,000. As the statute requires, the action was stayed and sealed while the government was notified of the information. The government declined to proceed with the action.2 Mathews decided to proceed with the action on the government’s behalf.
The district court then granted the Bank’s motion to dismiss her lawsuit with prejudice under the section of the Act establishing a jurisdictional bar where a claim is “based upon the public disclosure of allegations or transactions in a ... civil ... hearing ..., unless the ... person bringing the action is an original source of the information.” 31 U.S.C. § 3730(e)(4)(A). The court held that Mathews’ qui tam claim was “based upon” information which was “publicly disclosed” and that she was not an “original source” of the information. The interpretation and application of the language of § 3730(e)(4) are the main questions in her present appeal. We affirm, but for somewhat different reasons than those given by the court below.
I. Factual Background
Eunice Mathews personally guaranteed the extension of a $100,000 line of credit from the Bank of Farmington to her son Lester in 1981, thereby guaranteeing all of Lester’s then existing and future debts to the Bank. In 1987 and 1988, the Bank loaned Lester more than $290,000. All the loans were connected with Lester’s farming business. The Bank sought and obtained an FmHA guaranty for these loans, but, in violation of federal regulations, did not disclose in its application for the FmHA guaranty the existence of Mathews’ guaranty as collateral backing the new loans. Lester subsequently defaulted on the loans. The Bank submitted two loss reports to the FmHA in March 1993, neither of which disclosed the existence of Eunice Mathews’ guaranty. The FmHA paid out on these claims. The nondisclosure of the Mathews guaranty is the alleged violation of [857]*857the False Claims Act upon which she bases her qui tam claim. See 31 U.S.C. § 3729(1).
In June 1993, the Bank sued Mathews in Illinois state court to enforce her guaranty. She lost the suit on the instrument and lost her appeals, although the Bank settled for $100,000, waiving interest and attorney’s fees. In those proceedings, Mathews’ counsel, reviewing documents produced by the Bank in discovery, noticed that the existence of her guaranty had not been disclosed to the FmHA on the Bank’s applications to the agency. Under Illinois court rules, such discovery material is not filed in the public court file but is provided to opposing counsel. Her attorney attempted to use the Bank’s nondisclosure as a defense to the guaranty in the state trial. This effort failed, since no misconduct, illegality or crimes of the Bank against the government could affect Mathews’ own liability to the Bank under the guaranty.
The facts which constitute the basis of Mathews’ qui tam claim emerged in the state trial proceedings as follows. During discovery for those proceedings, as discussed, the Bank’s misrepresentation to the FmHA came to the attention of Mathews’ attorney when he received the Bank’s FmHA claims in discovery. In December 1993, a loan officer at the Bank, Mr. Rich Kimbrell, told Mathews’ attorney that the Mathews guaranty had not been disclosed on the FmHA application, but that it was in Lester’s file at the Bank and would have been reviewed periodically by the FmHA.
Mathews’ attorney then subpoenaed Mr. Victor Rhea, the FmHA employee principally responsible for the agency guaranty of Lester’s loans, for a deposition. Before the deposition, Rhea called Bank President Kent Kowal to inquire as to the reason for this subpoena. Kowal told Rhea that the Bank had sued Mathews on her guaranty. This explanation alerted Rhea to the existence of the Mathews guaranty. Rhea stated in an affidavit,in the qui tam case that this conversation was the first that he had heard of her guaranty. Since he should have been told about it from the beginning, this would have been an interesting conversation to have in detail, but the record in this case has no further information on its contents. Kowal provided an affidavit that confirms Rhea’s report. That is to say, the first time the FmHA heard of the Mathews guaranty was from the Bank itself, although this communication was instigated by Mathews’ subpoena. Kowal also stated in his, affidavit that he believes that he “advised” the FmHA about the Mathews guaranty during one of the previous conversations they had in regard to Lester’s loans, although he said that he did not recall a specific conversation with anyone in particular from the FmHA.
In the state, court litigation that followed, the Bank’s failure to disclose and the other facts described above became matters of publicly filed documentation and the subject of many court hearings. The Bank and the FmHA conducted separate negotiations about the Mathews guaranty. It appears from the record that the Bank paid FmHA some or all of the loss settlements that the Bank had received from FmHA in connection with Lester’s default. There is some dispute about precisely how much, but that is immaterial here.
In 1997, Mathews filed her qui tam claim in federal court, seeking $111,440.68 in damages. The question is whether, on these facts, a court has jurisdiction over this claim.
II. Discussion
A. Introduction
The False Claims Act was a Civil War statute, passed in 1863, originally to enable the federal government to punish and deter, the fraudulent claims of war profiteers. It provided criminal and civil penalties for presenting a false claim for payment against the United States. See S.Rep. No. 99-345, at 8 (1986), reprinted in 1986 U.S.C.C.A.N. 5266, 5273. From the very first, the statute included a qui tam provision.. The term comes from the Latin expression, qui tam pro domino rege quam pro se ipso in hac parte sequitur (“Who brings the action for the King as well as for himself’). A qui tam lawsuit is brought by a private party or “relator” who alleges fraud upon the government. If the claim is proven, the relator receives a percentage of the recovery rang[858]*858ing, under the current statute, from 10% to 30%. See 31 U.S.C. § 3730(d). The relator serves as a sort of private attorney general. The qui tam provision is based upon the idea “ ‘that one of the least expensive and most effective means of preventing frauds upon the Treasury is to make the perpetrators of them liable to actions by private persons acting ... under the strong stimulus of personal ill will or the hope of gain.’ ” United States ex rel. Marcus v. Hess, 317 U.S. 537, 541, n. 5, 63 S.Ct. 379, 383, 87 L.Ed. 443 (internal citations omitted).
The statute has been amended twice, once in 1943 and more recently in 1986. The original 1863 Act allowed anyone to bring a qui tam action and receive 50% of the amount recovered. S.Rep. No. 99-345, at 10. This broad provision led to abuse and in 1943, following Marcus, where the Supreme Court held that a relator could bring a qui tam action based entirely upon information contained in an indictment to which he had contributed nothing, Congress amended the statute to preclude actions “‘based on evidence or information the government had when the action was brought.’ ” See United States ex rel. Stinson v. Prudential Ins. Co., 944 F.2d 1149, 1153 (3d Cir.1991) (quoting 31 U.S.C. § 3730(b)(4) (1982) (superseded)). This led to claims being barred even in cases where the qui tam plaintiff supplied the information to the government before filing the claim. See United States ex rel. Wisconsin v. Dean, 729 F.2d 1100, 1106 (7th Cir.1984).
In 1986, Congress amended the Act again “to encourage any individual knowing of Government fraud to bring that information forward.” S.Rep. No. 99-345, at 2 U.S.Code Cong. & Admin.News 1986, pp. 5266-5267. The effect of these amendments on the whole is to broaden the qui tam provisions. Congress wanted to reward private individuals who take significant personal risks to bring wrongdoing to light, to break conspiracies of silence among employees of malfeasors, and to encourage whistleblowing and disclosure of fraud. See id. at 14. The legislative history shows that the 1986 amendments also, however, “‘sought to resolve the tension between ... encouraging people to come forward with information and ... preventing parasitic lawsuits.’ ” Cooper v. Blue Cross and Blue Shield of Florida, Inc., 19 F.3d 562, 565 (11th Cir.1994) (internal citations omitted). The provision the meaning of which is at issue in this case, § 3730(e)(4), establishes a “jurisdictional” bar against certain actions:
(e) Certain actions barred—
(4)(A) No court shall have jurisdiction over an action under this section based upon the public disclosure of allegations or transactions in a criminal, civil, or administrative hearing, in a congressional, administrative, or Government Accounting Office report, hearing, audit, or investigation, or from the news media, unless the action is brought by the Attorney General or the person bringing the action is an original source of the information.
(B) For purposes of this paragraph, “original source” means an individual who has direct and independent knowledge of the information on which the allegations are based and has voluntarily provided the information to the Government before filing an action under this section which is based on the information.
Congress intended that the courts not be troubled by persons who wish to capitalize on others’ discovery of frauds to the exposure of which they themselves have in no way contributed. However, the jurisdictional bar is not to be excessively narrowly construed. Our basic task in statutory interpretation is “to give effect to the intent of Congress.” United States v. American Trucking Ass’ns, 310 U.S. 534, 542, 60 S.Ct. 1059, 1063, 84 L.Ed. 1345. The provision is therefore to be understood in the context of the 1986 amendments, which, as explained, broaden the qui tam provisions, increasing incentives for the exposure of fraud.
A troubling aspect of Mathews’ claim turns on her attempt to recover from the Bank, in connection with an alleged false claim related to her guaranty, almost the precise amount she had to pay the Bank as a result of its state court lawsuit on her guaranty. This does not itself make her qui tam lawsuit opportunistic in the sense intended to [859]*859be precluded by the jurisdictional bar. The target of that provision is the sort of .opportunism involved in Marcus, where a relator recovered for reporting fraud that had been exposed entirely by others. But the purpose of the qui tam provision is to encourage and reward the exposure of fraud against the government, not to allow persons who have sustained financial liabilities backed by valid judgments or settlement orders to make those losses good at the expense of those to whom they are liable. Thus the court below is right that Mathews’ disclosure is not the kind of whistleblowing which the statute was designed to encourage. Nonetheless, had Mathews satisfied the jurisdictional requirements of 31 U.S.C. § 3730(e)(4), this troubling aspect of her lawsuit would not have been enough to create a jurisdictional bar. No such bar exists merely for lawsuits that are intended to make up for losses incurred in other lawsuits.
The inquiry into whether a court may hear a qui tam relator’s claim has three parts: (1) Have the allegations made by the plaintiff been “publicly disclosed”? (2) If so, is the lawsuit “based upon” that publicly disclosed information? (3) If so, is the plaintiff an “original source” of the information? Cooper, 19 F.3d at 564 n. 4. The threshold inquiry is whether there has been a public disclosure. Whether the plaintiff is an original source is immaterial unless there has been such public disclosure. See Wang v. FMC Corp., 975 F.2d 1412, 1416 (9th Cir.1992). The jurisdictional bar arises.only if the information upon which the qui tam claim is based has been publicly disclosed and the plaintiff was not the original source of the information. For an action to be barred under § 3730(e)(4), “each and every component of this section [must be] present.” Hinda v. University of Health Sciences/The Chicago Medical School, 65 F.3d 608, 613 (7th Cir.1995).
This Court reviews de novo a dismissal for lack of subject matter jurisdiction. See Health Cost Controls v. Skinner, 44 F.3d 535, 536 (7th Cir.1995); Joyce v. Joyce, 975 F.2d 379, 382 (7th Cir.1992). Mathews disputes that the statutory bar is properly construed as “jurisdictional.” It is not clear what she thinks is at stake in this theological dispute, since she concedes that she may be barred from maintaining the action on the grounds which Congress labeled as jurisdictional, whether or not they are in fact jurisdictional. She raises no issue of waiver, where the question of whether the bar is jurisdictional might be relevant. See Internat'l Union of Operating Engineers Local 150 v. Rabine, 161 F.3d 427, 429-432 (7th Cir.1998) (discussing the difference between lack of subject matter jurisdiction and failure to state a claim). But it is plain that in fact “satisfaction of § 3730(e)(4) is ... an issue of subject matter jurisdiction.” United States ex rel. Precision Co. v. Koch Indust., 971 F.2d 548, 551 (10th Cir.1992) (collecting cases). We have ourselves so held. See Houck on Behalf of United States v. Folding Carton Admin. Committee, 881 F.2d 494, 506 & n. 9 (7th Cir.1989). Mathews cites no caselaw to the contrary. The question presented, then, is whether dismissal for lack of subject matter jurisdiction was correct.
B. Analysis
1. Public Disclosure in a Hearing or Investigation
The court below concluded that the Bank’s failure to disclose the Mathews guaranty to the FmHA was “publicly disclosed” when the Bank released documents showing Mathews this failure in response to a discovery request. We agree that this information was publicly disclosed, but not for this reason. The court below, noting correctly that we have not spoken to whether discovery material, filed with the clerk of the court or not, is publicly disclosed within the meaning of § 3730(e)(4)(A),3 chose to follow the reasoning of Stinson, 944 F.2d at 1153. The Third Circuit held there that “disclosure of discovery material to a party who is not under any court imposed limitation as to its [860]*860use is a public disclosure under the [False Claims Act].” Stinson, 944 F.2d at 1158. The rationale for this is that “information disclosed in discovery is potentially accessible to the public.” Id. Whether information was filed is therefore not “significant.” Id. The Second Circuit adopted this approach without explanation in United States ex rel. Kreindler & Kreindler v. United Technologies Corp., 985 F.2d 1148, 1157 (2d Cir.1992). Relying on these cases, the court below concluded that since Mathews’ information was obtained in discovery, albeit unfiled discovery, it had therefore been publicly disclosed.
We think, however, that the reasoning of the Third Circuit is unsound. The interpretation of “public disclosure” adopted there runs contrary to the plain meaning of the words. “Absent a clearly expressed legislative intention to the contrary, th[e] language [of the statute itself] must ordinarily be regarded as conclusive.” Consumer Product Safety Comm’n v. GTE Sylvania, Inc., 447 U.S. 102, 108, 100 S.Ct. 2051, 2056, 64 L.Ed.2d 766. Words in a statute are to be given their plain and ordinary meaning. See United States v. James, 478 U.S. 597, 604, 106 S.Ct. 3116, 3121, 92 L.Ed.2d 483. “The plain language of a statute is the most reliable indicator of congressional intent.” Time Warner Cable v. Doyle, 66 F.3d 867, 876 (7th Cir.1995).
Here the language of the statute itself is “public disclosure,” not “potentially accessible to the public.” A plain and ordinary meaning of “public” is “open to general observation, sight, or cognition, ... manifest, not concealed,” 12 Oxford English Dictionary, 780 (2d ed. 1989) [hereinafter, OED]; that of “disclosure” is “opening up to view, revelation, discovery, exposure.” 4 OED, at 738. To say that something is publicly disclosed even if it is not in fact open to general observation or actually opened up to view, but is only potentially so, and that it is not publicly disclosed only if a court has forbidden its disclosure, is to distort the ordinary meaning of the words and in fact to read into the statute provisions that Congress did not enact.
Accordingly, we follow the D.C. Circuit, which held that “discovery material which has not been filed with the court and is only theoretically available upon the public’s request” is not “publicly disclosed” within the meaning of § 3730(e)(4)(A). United States ex rel. Springfield Terminal Ry. Co. v. Quinn, 14 F.3d 645, 652 (D.C.Cir.1994). As that court remarked, “[u]ntil discovery materials are filed with the court, we doubt that the discovery process conducted between two private litigants could itself constitute a public disclosure_[Unfiled discovery materials] are only potentially in the public eye.” Id. at 652-653.4 As the 10th Circuit observed in a different context, merely “conjectural or speculative ‘accessibility’ to the information [does not] bar[ ] the plaintiffs qui tarn action.” United States ex rel. Ramseyer v. Century Healthcare Corp. 90 F.3d 1514, 1521 (10th Cir.1996).5 We have previously held that a letter or other “communication between two private parties [themselves] is not public disclosure under this section.” [861]*861Hindo, 65 F.3d at 613 (citing Quinn, 14 F.3d at 652-653). There is no reason to treat unfiled discovery materials differently. To treat them the same best serves the policy of the False Claims Act, since barring actions based on information which was merely potentially but not actually opened up to view does not discourage parasitism. It only deters diligence in uncovering fraud.
Nonetheless, the information about the Bank’s misrepresentations upon which Mathews bases her qui tarn action was indeed publicly disclosed because the Bank had disclosed it to a competent public official, here Victor Rhea of the FmHA. Disclosure of information to a competent public official about an alleged false claim against the government we hold to be public disclosure within the meaning of § 3730(e)(4)(A) when the disclosure is made to one who has managerial responsibility for the very claims being made. This construction accords with a standard meaning of “public,” which can also be defined as “authorized by, acting for, or representing the community.” 12 OED, at 779. Disclosure to an official authorized to act for or to represent the community on behalf of government can be understood as public disclosure.
The point of public disclosure of a false claim against the government is to bring it to the attention of the authorities, not merely to educate and enlighten the public at large about the dangers of misappropriation of their tax money. Disclosure to the public at large is a step in lowering the jurisdictional bar precisely because it is likely to alert the authorities about the alleged fraud. After investigation, they can take the proper steps to deal with it — prosecution, settlements involving repayment of funds, or whatever may be called for in the particular case. Since a public official in his official capacity is authorized to act for and to represent the community, and since disclosure to the public official responsible for the claim effectuates the purpose of disclosure to the public at large, disclosure to a public official with direct responsibility for the claim in question of allegations or transactions upon which a qui tarn claim is based constitutes public disclosure within the meaning of § 3730(a)(4).
Disclosure to officials with less direct responsibility might still be public disclosure if the disclosure is public in the commonsense meaning of the term as “open” or “manifest” to all. See United States ex rel. Fine v. Chevron, Inc., 72 F.3d 740, 743 (9th Cir.1995) (information contained in Inspector General’s semi-annual report to Congress publicly disclosed). The more open a disclosure is, the less any public official need be specifically informed. If it is sufficiently open, no official need be specifically informed. The more likely the competent official is to be apprised of the relevant facts by a disclosure, the less “public,” in the sense of open or manifest to all, it need be.6 If the disclosure is made, as here, to precisely the public official responsible for the claim, it need not be disclosed to anyone else to be public disclosure within the meaning of the Act.
The disclosure, if not actually made to the public at large, must be to a public official. Private persons do not represent the public. Mathews’ case is distinguishable from one where the results of an internal corporate investigation not disclosed within the corporation nor disseminated to the public, “d[id] not trigger the public disclosure bar .... ” United States ex rel. Aflatooni v. Kitsap Physicians Services, 158 F.3d 439, 446 (9th Cir.1998).
Clearly, however, not all disclosure to a public official is public disclosure. Assuming no other public promulgation of the information, the public official to whom the information is disclosed must be one whose duties extend to the claim in question in some significant way. It would not have been public disclosure here had the Bank divulged the information in this case to a postal carrier or to the Governor of Guam and to no one else. Accordingly we reject the construction adopted by the 10th Circuit, according to which there is public disclosure if “the allegations [are disclosed] to any [single] member of the public not previously [862]*862informed thereof,” United States ex rel. Fine v. Advanced Sciences, Inc., 99 F.3d 1000, 1006 (10th Cir.1996), even to a postal carrier or to the Governor of Guam — or (as in Advanced Sciences), to a representative of the American Association of Retired Persons. That interpretation fails to serve the purpose of the Act under the 1986 amendments, namely to encourage the exposure and punishment of fraud.
It should be obvious that disclosure to the government by a qui tam plaintiff of the information on which the complaint is based, as mandated in § 3730(a)(2), is not “public disclosure” under § 3730(e)(4)(A), or the jurisdictional bar would prohibit any qui tam action whatsoever by someone who was not an original source, which is clearly not the intent of Congress.
In this case, however, the information was disclosed to Victor Rhea, the FmHA official responsible for the very loans Mathews had, hitherto unbeknownst to him, guaranteed. It may not have been the Bank’s intent to confess or correct its own misrepresentation, but the effect of its action was to bring just that misrepresentation to the attention of precisely the person who should have been apprised of the information to begin with, thus effectuating the purpose of public disclosure.
The jurisdictional bar and the “original source” exception operate only when the information upon which a qui tam claim is based is publicly disclosed in a “criminal, civil, or administrative hearing, in a congressional, administrative, or Government Accounting Office report, hearing, audit, or investigation, or from the news media.” 31 U.S.C. § 3730(e)(4)(A). The court below, treating the disclosure as having taken place in discovery, addressed whether discovery is encompassed in the statutory term “hearing.” Several Circuits have so held. See, e.g., Quinn, 14 F.3d at 652 (“For purposes of § 3730(e)(4)(A), ‘hearing’ is roughly synonymous with ‘proceeding.’ ”); accord Kreindler, 985 F.2d at 1158; Stinson, 944 F.2d at 1155-1157; Siller, 21 F.3d at 1350-1351.
We need not reach the question here, however, because the “hearing” clause is not the applicable provision in the present case. The FmHA learned of the Mathews guaranty and the Bank’s misrepresentation because of a step in the discovery proceedings, but not, to use the statutory language “in” a hearing, even if discovery proceedings are hearings. The FmHA’s first inkling of these facts (at the latest) was the phone conversation during which Bank President Kowal told FmHA official Rhea of the Bank’s guaranty suit against Mathews. Even if “in” is as flexible as “hearing,” an informal conversation between a potential witness and a party’s agent about a deposition subpoena is not itself part of any proceeding, and so not “in a hearing.”
The statutory language that brings the “public disclosure” raised in this conversation within the scope of § 3730(e)(4)(A) is that the bar operates against public disclosure in “a congressional [or] administrative ... investigation,” id. The conversation between Rhea and Kowal was pursuant to an administrative investigation. An official of an administrative agency faced with an anomaly (the subpoena) in a matter within his purview made an inquiry to an official of a regulated industry for which the agency was responsible. It does not stretch the language to understand such “inquiry” as an “investigation.” Although there seems to be no caselaw discussing what is necessary for something to be an administrative investigation within the meaning of § 3730, investigations need not be as formal as the “multifaceted investigation” in United States ex rel. John Doe v. John Doe Corp., 960 F.2d 318, 323 (2d Cir.1992). They may be informal or casual inquiries so long as they are undertaken by authorized officials with official purposes. See, e.g., O’Connor v. Chicago Transit Authority, 985 F.2d 1362, 1364 (7th Cir.1992) (informal administrative inquiry by whistleblower characterized as an investigation). A police officer, hearing a peculiar noise in a dark shop, investigates by gingerly shining a flashlight inside and asking, “What’s up?” This is essentially what Mr. Rhea did in his phone call. The allegations or transactions which constitute the basis of Eunice Mathews’ claim were then “publicly disclosed” in an “administrative ... investigation.”
[863]*8632. “Based Upon” Publicly Disclosed Information
The jurisdictional bar applies only when a qui tam action is “based upon” a public disclosure of the allegations or transactions. The majority view is that of the Second Circuit, that a qui tam action is “based upon” a public disclosure when the supporting allegations are “the same as those that have been publicly disclosed .... regardless of where the relator obtained his information.” Doe, 960 F.2d at 324; accord Kreindler, 985 F.2d at 1158; Quinn, 14 F.3d at 652-655 (D.C.Cir.); McKenzie, 123 F.3d at 940 (6th Cir.); Wang, 975 F.2d at 1417 (9th Cir.); Koch, 971 F.2d at 552 (10th Cir.); Cooper, 19 F.3d at 566-567 (11th Cir.). The rationale in these cases is that the jurisdictional bar operates if the information upon which the suit is based is already “ ‘in the public domain, and the qui tam plaintiff was not a source of that information,’ ” Doe, 960 F.2d at 324 (internal citations omitted). Under this reading, once the Bank had informed the FmHA of the facts surrounding its misrepresentation, Mathews’ claim was “based upon” that disclosure even if she knew about those facts entirely apart from the Bank’s admission, e.g., through examination of the documents produced in discovery.
The Fourth Circuit, representing the minority view, objects that the statutory language “based upon” does not mean “similar (even identical) to” but “derived from.” See Siller, 21 F.3d at 1347-1348. On this construction, it is irrelevant for purposes of the jurisdictional bar that Mathews’ claim- states the same information that the Bank divulged to the government unless Mathews got the information on which her claim is based from the Bank’s divulgence. The rationale of this reading is twofold. One is textual: “based upon” is not synonymous with “identical to” or “similar to,” but can be substituted for “derived from” wherever it occurs. The other rationale is policy: a lawsuit based upon information which happens to be similar or identical to publicly disclosed allegations or transactions, but which derives from some other source than the public disclosure, is not parasitic, and should not be barred by a provision meant to bar parasitic lawsuits. Id. at 1348.
The Fourth Circuit’s interpretation of “based upon” is the better on the grounds both of plain meaning and public policy. Mathews’ claim, however, is “based upon” a public disclosure even under our preferred “derived from” reading. Her complaint and pleadings in fact refer not only to the existence of the Mathews guaranty and its nondisclosure on the Bank’s submissions to the FmHA, but also, and essentially, to Rhea’s statement that he had not learned of the guaranty prior to Mathews’ subpoena and his own conversation with the Bank. Whether or not the unfiled discovery material upon which Mathews in part bases her lawsuit constitutes a public disclosure, Rhea’s conversation with Kowal, the latest point at which the Bank divulged the existence of the guaranty to the government, constitutes a public disclosure in the context of an administrative inquiry. Mathews’ qui tam claim is actually and substantially derived from this disclosure. It is part of her claim, cited in her complaint, and necessary to show that the Bank had not corrected its misrepresentation to the FmHA until Mathews began discovery in the state court action. This is sufficient for her claim to be “based upon” the disclosure for purposes of the jurisdictional bar.
If the public disclosure from which the information is actually derived is essential to a qui tam claim, then the claim is based upon the public disclosure for the purposes of the jurisdictional bar. This Court need not reach whether § 3730(e)(4)(A) bars qui tam actions based “in any part” rather than “solely” upon publicly disclosed information. In Houck, we said that “the Act ... does not allow a qui tam plaintiff to bring an action based solely upon publicly disclosed transactions.” Houck, 881 F.2d at 504. This does not say, however, that the jurisdictional bar only applies where a plaintiffs information is based wholly upon public disclosures. It is consistent with saying that a qui tam plaintiff may not bring an action based in any part on such transactions. See, e.g., Kreindler, 985 F.2d at 1158 (“solely” not part of the statute); accord Koch, 971 F.2d at 553 (“Based upon” means “based upon in any [864]*864part.”)- With the qualification noted below, we need not decide the broader question. The more limited holding will suffice here, that a claim which both depends essentially upon publicly disclosed information and is actually derived from such information is “based upon” a public disclosure for purposes of § 3730(e)(4)(A). Mathews’ claim is therefore “based upon” a public disclosure.
3. Original Source
Since Mathews based her qui tam action upon information publicly disclosed in (here) an administrative investigation, she must be an “original source” of that information to escape from the jurisdictional bar. According to the statute:
“[OJriginal source” means an individual who has direct and independent knowledge of the information on which the allegations are based and has voluntarily provided the information to the Government before filing an action under this section which is based on the information.
31 U.S.C. § 3730(e)(4)(B). In one of the earliest decisions construing this section, Houck, 881 F.2d 494, we construed the expression “original source” to mean that, to qualify as such a source, a qui tam plaintiff had to satisfy two requirements. The plaintiffs knowledge of the information upon which the claim was based had to be “direct,” a term we did not define but indicated was satisfied by the Houck plaintiffs involvement in assisting claimants with the filings of the claims from a fund whose activities he alleged raised statutory concerns under the False Claims Act. Id. at 504-505. The plaintiffs knowledge also had to be “independent,” which we defined as independent of the public disclosure, saying in that case: “There is no evidence to indicate that Houck would have learned of the claims to the funds absent public disclosure.” Id.
Under the Houck standard, Mathews is not an original source, because, whether or not her knowledge is direct, it is not independent of the public disclosure. As explained, her claim is actually and substantially based upon the Bank’s disclosure to the FmHA of the guaranty in the context of Rhea’s investigation. Mathews’ knowledge of a piece of information essential to her claim, that the Bank had not divulged this information to the FmHA prior her discovery request in the state lawsuit, entirely derives from or depends upon Rhea’s testimony, later supported by Kowal, that the FmHA learned about the existence of the Mathews guaranty, and thus the Bank’s previous misrepresentations, in Rhea’s investigation. But since her knowledge depends on Rhea’s testimony as to his findings in the investigation, that knowledge is not independent of the investigation, and therefore is not independent of the public disclosure.
Mathews argues that she is an original source under Houck because, while some pieces of her story of fraudulent conduct by the Bank may have been dependent on public disclosures, her contribution was to reveal the entire pattern of the fraud. All the facts from various sources came together in her lawyer’s office, she says, before she caused the facts to be placed in the public domain. The hidden pattern she revealed is what was independent of public disclosure. Fair enough — in theory. In an exceptionally or unusually complicated allegation of fraud each piece of the information may be publicly disclosed, yet the fraud itself may remain hidden until some perspicacious plaintiff puts it in perspective. We acknowledge that in such a case, a plaintiff might be an original source even though her knowledge of every isolated element of the fraud is based upon public disclosures. Mathews, however, is not such a plaintiff.
The fraud alleged is a simple affair, revealed completely by one line on each of three pieces of paper — the Bank’s applications to the FmHA stating that the value of any personal guarantee was $0.0 — and one other true statement that such a guaranty existed, like that Kowal made when Rhea called him to ask about the subpoena. This was no profound scheme. It would not take Sherlock Holmes to figure it out. Kowal’s response would have revealed the misrepresentation to any sentient FmHA official. We cannot say in advance how complex a fraud must be, or how deep or clever its revelation, to satisfy the original source requirement when the claim is based on information which [865]*865has been publicly disclosed. But the misrepresentation Mathews revealed, and the difficulty involved in its unveiling, falls well short of the mark. Putting “two and two together,” as Mathews says she did, will not do. To qualify under the original source provision solely by putting things together one has to be doing more than simple arithmetic.
Although Mathews does not qualify as an original source under Houck, it has been over nine years since we decided that case. We briefly consider whether subsequent caselaw suggests revisiting our earlier view. The Ninth Circuit accepts our conjunctive reading of “original source” but adds a requirement that the qui tam plaintiff herself “have had a hand in the public disclosure of allegations that are part of one’s suit.” Wang, 975 F.2d at 1418; accord United States ex rel. Dick v. Long Island Lighting Co., 912 F.2d 13, 16 (2d Cir.1990). This view is rejected as having no basis in the text or legislative history by Stinson, 944 F.2d at 1160 (3d Cir.); Siller, 21 F.3d at 1355 (4th Cir.); Advanced Sciences, 99 F.3d at 1006-1007 (10th Cir.), and Cooper, 19 F.3d at 568 n. 13 (11th Cir.). We so agree. The statute says that the jurisdictional bar operates when a qui tam claim is based upon publicly disclosed allegations or transactions “unless ... the person bringing the action is an original source of the information.” 31 U.S.C. § 3730(e)(4)(A) (emphasis added). It does not say “is an original source of the public disclosure.” But this does not help Mathews. The problem is not that she was not the source of the disclosure, although she was not, but that she was not an original source of the information, someone who would have learned of the allegation or transactions independently of the public disclosure.
To be an “original source,” a qui tam plaintiff must be a source as well as being an original source. The language of 31 U.S.C. § 3730(e)(4)(B) (quoted supra), defining “original source,” has two requirements. To be “original” the plaintiff must have “direct and independent knowledge of the information on which the allegations are based.” To be a “source” the plaintiff must have “voluntarily provided the information to the Government before filing an action .... ” Id. Granted, to say that a “source” is one who voluntarily provides the government with information before filing a qui tam action is a nonstandard definition. But it is hard to make any other sense of § 3730(e)(4)(B), which expressly tells us that that section says what “original source” means for the purposes of the qui tam provision of the False Claims Act. The subsection defines a two-word term with two parallel requirements. The requirement of “direct and independent knowledge” obviously defines “original,” since someone might have such knowledge and not be a source. Thus “providing the information to the government before filing the action” must be a construction of “source,” and we so read it. “In the absence of statutory definition, we give terms their ordinary meaning,” Bass v. Stolper, Koritzinsky, Brewster & Neider, S.C., 111 F.3d 1322, 1325 (7th Cir.1997), but when there is an express statutory definition, even if an odd, nonstandard, or technical definition, we are bound to follow it. ‘“Courts must presume that a legislature says in a statute what it means and means in a statute what it says there.’ ” In the Matter of Lifschultz Fast Freight Corp., 63 F.3d 621, 628 (7th Cir.1995) (quoting Connecticut Nat. Bank v. Germain, 503 U.S. 249, 253-254, 112 S.Ct. 1146, 1149, 117 L.Ed.2d 391).
We feel confident in construing the term “source” in this manner because several other circuits read the language as we do. The D.C. Circuit has said that “[t]o qualify as an ‘original source,’ a relator must also have ‘voluntarily provided information to the government’ before filing a qui tam suit which is ‘based on the information.’” Findley, 105 F.3d at 690; accord McKenzie, 123 F.3d at 942 (“A ‘true whistleblower’ [must] ... alert[ ] the government to the alleged fraud before such information is in the public domain.”); United States ex rel. Barth v. Ridgedale Elec., Inc., 44 F.3d 699, 704 (8th Cir.1995). These courts do not expressly say that they are. interpreting the term “source,” but we understand them to be doing so. In any case the functional effect of their constructions is the same as. ours. To escape the jurisdictional bar, a plaintiff must, before [866]*866filing her qui tam action, have voluntarily-provided the information upon which her qui tam claim is based to the government if the information has been publicly disclosed.
Precisely what it would mean to have voluntarily provided the information to the government before filing the lawsuit under 31 U.S.C. § 3730(e)(4)(B) is not something settled in the caselaw or clearly specified in the statute. A qui tam plaintiff might satisfy this requirement, for example, by notifying the United States Attorney, the FBI, or other suitable law enforcement office of the information which is the basis for the action, or by informing the agency or official responsible for the particular claim in question&emdash;in Mathews’ ease, the FmHA or Victor Rhea. These are not exclusive methods, but it is clear that the requirement is not satisfied by informing the government at the time of filing the action, even in compliance with the requirement that a private plaintiff must provide the government, at the time of filing, with “a copy of the complaint and written disclosure of substantially all material evidence and information the person possesses _"Id. § 3730(b)(2). More must be done to qualify as an original source than to file the action. The government must be voluntarily notified beforehand. See Findley, 105 F.3d at 690-691.
In this case, nothing in the record indicates that Mathews did anything to voluntarily provide the information on which her allegations are based to the government before filing her lawsuit. In the circumstances, then, she fails to qualify as a source, much less as an original source. The “source” of the government’s information about the Bank’s misrepresentations was in fact the Bank itself. It was Bank President Kowal who voluntarily provided FmHA official Rhea with the information about the existence of the Mathews guaranty in the course of Rhea’s investigation. True, Mathews’ subpoena caused Rhea to begin this investigation by telephoning Kowal. But the jurisdictional provision of the statute requires that a qui tam plaintiff who bases her case upon publicly disclosed allegations or transactions be an original source, not an original cause. Her position in the causal chain is insufficient to create jurisdiction. Since Mathews is not a source, much less an original source, the requirements for jurisdictional bar are satisfied. By the terms of the statute, no court has jurisdiction over her claim.
A final remark seems in order. It might seem that Mathews loses because of odd quirks of timing. Had Kowal not informed Rhea of the Mathews guaranty when Rhea called to investigate the circumstances around Mathews’ subpoena, then the information might not have been publicly disclosed. Had Mathews initiated contact with Rhea about the Bank’s misrepresentation before Rhea called Kowal, Mathews’ lawsuit might not be based upon public disclosures, since there would have been no investigation, merely a private conversation. Had Mathews informed Rhea of the Bank’s misrepresentation before the Bank did, she might be the original source.
Should jurisdiction depend upon such quirks? Do we really want a rule which says: Get your qui tam claim in quickly? The short answer is yes, because that is the rule Congress adopted. Where the statute makes jurisdiction depend on events which occur at determinable times, such as a public disclosure of information or its voluntary provision to the government before filing a lawsuit, a plaintiff is encouraged not to dawdle. Just as one can lose a right to sue by the running of a statute of limitations, so a court can be denied jurisdiction by such an accident of timing. But the policy rationale is clear: The “intent of the Act ... is to encourage private individuals who are aware of fraud against the government to bring such information forward at the earliest possible time and to discourage persons with relevant information from remaining silent.” Barth, 44 F.3d at 704 (internal citations omitted) (emphasis added); see also Wang, 975 F.2d at 1419; Cooper, 19 F.3d at 565. These goals are promoted by a jurisdictional rule requiring early divulgence of allegations of fraud.
AFFIRMED.