concurring/dissenting.
I join Parts I, 11(A), 11(C)(1), 11(C)(3), II(D), TV, and VI of Judge Lucero’s opinion. I concur in the result in Parts 11(B) and 11(C)(2). I dissent from Part III.
I. Special Concurrence Regarding Part 11(B) (Regence Immunity)
I agree that Regence is not immunized by 42 U.S.C. § 1395u(e) (1994) from all liability arising from its payment of claims. As I shall try to explain, I think that the most reasonable reading of § 1395u(e)(3) is that it immunizes the carrier from liability for an employee’s act when the employee is immunized from liability for that act by paragraph (1) or (2) of § 1395u(e).
Section 1395u(e) states:
(1) No individual designated pursuant to a contract under this section as a certifying officer shall, in the absence of gross negligence or intent to defraud the United States, be liable with respect to any payments certified by him under this section.
(2) No disbursing officer shall, in the absence of gross negligence or intent to defraud thé United States, be liable with respect to any payment by him under this section if it was based upon a voucher signed by a certifying officer designated as provided in paragraph (1) of this subsection.
(3) No such carrier shall be liable to the United States for any payments referred to in paragraph (1) or (2).
The thrust of subsection (e) is to eliminate suits based on errors committed without gross negligence or fraudulent intent. Paragraph (1) says that the certifying officer is not liable for certifying a payment unless the officer acted with gross negligence or fraudulent intent. Paragraph (2) states that the disbursing officer is not liable for making a certified payment unless the officer acted with gross negligence *732or fraudulent intent. Then, to avoid suits that skip the middleman and go directly against the carrier, paragraph (3) immunizes the carrier from liability when its employee is immune.
To be sure, paragraph (3)’s language poses difficulties. For one thing, it is unclear what the antecedent of such is in the phrase “such carrier.” Much more importantly, the phrase “any payments referred to in paragraph (1) or (2)” is ambiguous. Regence would have us read the phrase as encompassing all payments, or at least all payments certified by a certifying officer. This is a possible reading. After all, paragraphs (1) and (2) describe payments certified by a certifying officer for which a certifying or disbursing officer may be liable (when the certifying or disbursing officer acts with gross negligence or intent to defraud) and payments certified by a certifying officer for which the certifying or disbursing officer is immune from liability (when there is no such gross negligence or fraudulent intent). Thus, paragraphs (1) and (2) can be said to “refer to” certified payments for which the officers are immune and certified payments for which they are not — that is, all certified payments. Under Regenee’s reading, carriers would be immune under paragraph (3) from liability for all payments certified by a certifying officer.
But if the intent of paragraph (3) — “No ... carrier shall be liable ... for any payments referred to in paragraph (1) or (2)”- — -were to immunize carriers for all payments certified by certifying officers, one wonders why the drafters chose such a peculiar way to say it. A more natural mode of expression would have been, “no carrier shall be liable for any payments certified by a certifying officer.” And if the “payments referred to in paragraph (1) or (2)” are both payments for which an officer may be liable and payments for which an officer is immune, why include “or (2)” at the end of the quoted phrase? Nothing would be lost by saying merely “any payment referred to in paragraph (1),” because — if one says that each paragraph “refers to” payments for which there may be liability as well as payments for which there is immunity' — -the same payments are “referred to” in both paragraphs (1) and (2). Each paragraph, under Regence’s reading, addresses all certified payments: For each payment, the certifying officer is either subject to liability or immune, and the same goes for the disbursing officer; paragraph (1) addresses the certified payments for which certifying officers are immune or may be liable (that is, all certified payments), and paragraph (2) addresses certified payments for which disbursing officers are immune or may be liable (that is, all certified payments). Thus, in the phrase “payments referred to in paragraph (1) or (2),” the words “or (2)” are surplusage. Ordinarily, we should avoid a construction of a statute that renders portions of the statutory language superfluous. See Arlington Cent. Sch. Dist. Bd. of Educ. v. Murphy, — U.S. -, - n. 1, 126 S.Ct. 2455, 2460 n. 1, 165 L.Ed.2d 526 (2006) (but noting that “instances of surplusage are not unknown”).
I would interpret paragraph (3) differently. First, one must read § 1395u(e) in context. That context is liability for erroneous payments. It makes no sense to provide immunity unless the immunized conduct may otherwise generate liability. Correct payments do not generate liability. The risk of liability arises only when a certifying or disbursing officer, through fault (negligence or otherwise), does something leading to an erroneous payment. Accordingly, when paragraph (1) says that a certifying officer shall not be liable “with respect to any payments certified by him under this section,” it is implicitly refer*733ring only to payments for which the officer would otherwise be liable — that is, payments based on certifications that were erroneous because of the officer’s fault. Immunity would be unnecessary with respect to any other payments.
The purpose of paragraph (1), then, is to carve out from the set of payments for which the certifying officer may be liable those payments for which the officer is immune. Those payments are payments that were erroneously made because of the certifying officer’s fault, but when the fault did not rise to gross negligence or intentional fraud. Given that purpose, it is natural to say that the payments “referred to in paragraph (1)” are the carved-out payments, those for which the certifying officer might have been liable (because of fault) but for the immunity provided when the officer did not act with gross negligence or fraudulent intent.
Similarly, the payments “referred to in paragraph ... (2)” are the certified payments for which the disbursing officer might have been liable but for his statutory immunity. And paragraph (3)’s protection of carriers from liability for “any payments referred to in paragraph (1) or (2)” therefore provides essentially respondeat-superior immunity. If the carrier would otherwise be liable for an erroneous certified payment because of the fault of a certifying or disbursing officer, the carrier is immune when the officer did not act with gross negligence or intent to defraud.
Accordingly, I join the majority in rejecting Regence’s defense that it is immune under § 1395u(e) from Sikkenga’s claim.
II. Special Concurrence Regarding Part 11(C)(2) — Causation
I also agree with the majority that Sik-kenga’s complaint states a claim that Re-gence caused ARUP to submit a false claim. In my view, however, we should be wary of applying tort concepts of causation to the False Claims Act because of its long-term congruence with a criminal statute and its present punitive provisions.
I begin with some history of the False Claims Act. The original 1863 Act was a criminal statute which included a provision for civil claims. Section 1 imposed a criminal penalty on military personnel who “presented] or cause[d] to be presented for payment or approval ... any claim upon or against the Government of the United States ... knowing such claim to be false, fictitious, or fraudulent.” (I fail to see a material difference from the present language of 31 U.S.C. § 3729(a): “knowingly presents, or causes to be presented ... a false or fraudulent claim for payment or approval.”) Section 3 of the 1863 statute applied to nonmilitary personnel, imposing civil liability (double damages and a $2,000 penalty) on those “who shall do or commit any of the acts prohibited by any of the foregoing provisions,” as well as criminal punishment if convicted. In 1874 the criminal provisions of former sections 1 and 3 were consolidated in Revised Statutes of the United States, Title 70 (Crimes) § 5438, while the civil provisions were moved to Title 36 (Debts Due by or to the United States). But the civil provisions still cross-referenced the criminal statute for a description of the prohibited conduct. See Rev. Statutes, Title 36 § 3490 (“any [nonmilitary] person ... who shall do or commit any of the acts prohibited by any of the provisions of [§ 5438] shall forfeit [$2,000 plus double damages]”). Although § 5438 was repealed in 1909 and replaced by two separate criminal statutes, the repealed statute “ha[d] continued vitality ... insofar as it specified] the acts giving rise to civil liability under § 3490.” United States v. Bornstein, 423 U.S. 303, 305 n. 1, 96 S.Ct. 523, *73446 L.Ed.2d 514 (1976). Only in 1982 did Congress enact legislation that made the False Claims Act’s civil provisions freestanding, without a cross-reference to a criminal statute. See Pub.L. No. 97-258, § 3729, 96 Stat. 877, 978 (1982).
Courts strictly construe ambiguous language in criminal statutes in favor of lenity. See United States v. Kozminski, 487 U.S. 931, 952, 108 S.Ct. 2751, 101 L.Ed.2d 788 (1988). This rule of strict construction applies even when the language is applied in a civil context. See Crandon v. United States, 494 U.S. 152, 158, 110 S.Ct. 997, 108 L.Ed.2d 132 (1990) (“[B]ecause the governing standard is set forth in a criminal statute, it is appropriate to apply the rule of lenity in resolving any ambiguity in the ambit of the statute’s coverage.”). Thus, at least through 1982 the civil provisions of the False Claims Act were to be construed strictly. Indeed, United States v. McNinch, 356 U.S. 595, 78 S.Ct. 950, 2 L.Ed.2d 1001 (1958), in holding that an application for credit insurance was not a “claim” within the meaning of the False Claims Act, stated: “[I]t must be kept in mind ... [that] we are actually construing the provisions of a criminal statute. Such provisions must be carefully restricted, not only to their literal terms but to the evident purpose of Congress in using those terms, particularly where they are broad and susceptible to numerous definitions,” id. at 598, 78 S.Ct. 950 (internal footnote omitted). To be sure, the civil and criminal provisions have since then been technically divorced; but the pertinent language of the civil provision has not materially changed, so there is no reason to believe that the language should be interpreted any differently now than it should have been in 1909 (or 1982). See Cook County v. United States ex rel. Chandler, 538 U.S. 119, 132, 123 S.Ct. 1239, 155 L.Ed.2d 247 (2003) (refusing to infer that 1986 amendments to False Claims Act silently redefined the word person in the statute).
Reinforcing this view is that the False Claims Act is a punitive statute, and civil punitive statutes, like criminal statutes, are to be construed strictly. See Comm’r v. Acker, 361 U.S. 87, 91, 80 S.Ct. 144, 4 L.Ed.2d 127 (1959). The Act is punitive in two respects. The availability of treble damages, even though it has “a compensatory side,” Cook County, 538 U.S. at 130, 123 S.Ct. 1239, also has a punitive character, see Vt. Agency of Natural Res. v. United States ex rel. Stevens, 529 U.S. 765, 785, 120 S.Ct. 1858, 146 L.Ed.2d 836 (2000). In addition, § 3729(a)(7) of the Act provides for a penalty of $5,000 to $10,000 regardless of actual damages. See Wood, Walker & Co. v. Evans, 461 F.2d 852, 855 (10th Cir.1972) (courts strictly construe statutes under which “the amount of the damages is fixed in a somewhat liquidated measure without regard to injury suffered”).
Accordingly, I would refrain from “bor-rowfing] traditional principles of tort law to analyze causation for damages under the FCA.” Op. at 23. At this stage of the case, however, it is unnecessary to explore the precise scope of causation under the False Claims Act. We are reviewing a dismissal for failure to state a claim. The allegations of the Complaint are therefore taken as true. Sikkenga alleges that any claim submitted to Regence with diagnosis code 796.4 is a false claim. I confess to some skepticism about the allegation. I would think that a claim submitted with an improper code is simply an improperly documented claim; the underlying claim may still be proper, and compensable once the documentation is corrected. But Re-gence has not challenged the allegation, perhaps because it is a matter to be decided after the presentation of evidence, not when ruling on the pleadings. As for cau*735sation, Sikkenga’s complaint explicitly alleges that Regence “caused ARUP to present false claims for payment or approval.” ApltApp. at 489 (Am. Compl. at ¶ 140). Moreover, Sikkenga describes the manner of causation. She alleges that Regence told ARUP that it would accept claims submitted with a 796.4 code. Given the ongoing relationship between Regence and ARUP, for Regence to inform ARUP that it would process claims that are false on their face could, in my view, constitute causing, in the criminal-law sense, ARUP to submit false claims. The typical context in which “causing” an act to occur arises in criminal cases is when the act is performed by an innocent party rather than by a partner in crime. For example, in a mail-fraud case the defendant “causes” a mailing to take place by handing the envelope to a friend or secretary to take to the post office, or by depositing an out-of-state check in a bank (which, at least in the old days, would need to mail the check to the originating bank for clearance). See, e.g., Pereira v. United States, 347 U.S. 1, 8-9, 74 S.Ct. 358, 98 L.Ed. 435 (1954). What the defendant does is to set in motion a process that in the ordinary course will result in the prohibited action, without the need for any intermediary to have a nefarious motive. Sikkenga’s complaint appears to encompass an allegation that Regence in this sense caused the filing of false claims by ARUP' — perhaps ARUP had a nefarious motive, but false claims would have been submitted regardless of that motive.
I should add, however, that I doubt that causation could be shown by evidence that Regence said only that it would accept claims (that may or may not be legitimate) without adequate documentation substantiating that they are proper. In that circumstance, the submitter of the claims is not being advised to submit false claims, only that it is being trusted not to do so.
III. Dissent Regarding Part III (Statute of Limitations)
Finally, I disagree with the majority’s construction of the False Claims Act statute of limitations, 31 U.S.C. § 3731(b). The statute reads:
(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.
I join the majority in rejecting the Ninth Circuit’s view in United States ex rel. Hyatt v. Northrop Corp., 91 F.3d 1211 (9th Cir.1996), that a relator can be “the official of the United States charged with responsibility to act” in paragraph (2). But I cannot agree that the statute can be read to say that paragraph (2) does not apply to suits by relators. Our view of what Congress must have intended cannot substitute for statutory language. See Robbins v. Chronister, 435 F.3d 1238, 1241 (10th Cir.2006) (en banc) (strictly limiting use of absurdity doctrine to construe statute contrary to its language). Congress may have wanted to limit relators to the six-year limitation period; but it did not say so. I agree with District Judge Benson that there is no ambiguity to resolve. See United States ex rel. Colunga v. Hercules, Inc., No. 89-CV-954B, 1998 WL 310481, *5 (D.Utah Mar.6, 1998). The majority’s invocation of the absurdity doctrine makes no attempt to establish the *736satisfaction of the extremely strict conditions for application of that doctrine set forth in our recent unanimous en banc opinion on the subject. See Robbins. In any event, as Judge Benson explained, the plain meaning of § 3731(b) is not absurd. Congress could well have decided that a relator should not be time-barred if the government is not. To bar the relator but not the government may accomplish nothing more than preventing the relator from securing her just reward in bringing the matter to court. I would not foreclose, however, the possibility that an equitable doctrine implicitly incorporated in the statute could bar a relator who delays, for improper reasons, reporting fraud to the Government. Cf. Young v. United States, 535 U.S. 43, 49, 122 S.Ct. 1036, 152 L.Ed.2d 79 (2002) (noting that equitable tolling is a background principle generally applied in construing statutes of limitations).