FILED
                                                        United States Court of Appeals
                                                                Tenth Circuit

                                                            September 4, 2012
                                      PUBLISH              Elisabeth A. Shumaker
                                                               Clerk of Court
                     UNITED STATES COURT OF APPEALS

                                 TENTH CIRCUIT


In re: KENNETH WOOLSEY;
STEPHANIE WOOLSEY,

                Debtors.

------------------------------

KENNETH WOOLSEY; STEPHANIE
WOOLSEY,

                Appellants,

v.
                                                      No. 11-4014
CITIBANK, N.A.,

                Appellee,

KEVIN R. ANDERSON, Chapter 13
Trustee,

                Trustee - Appellee,

NATIONAL ASSOCIATION OF
CONSUMER BANKRUPTCY
ATTORNEYS,

                Amicus Curiae.


                   Appeal from the United States District Court
                             for the District of Utah
                         (D.C. No. 2:10-CV-01097-BSJ)
David M. Cook, Salt Lake City, Utah, for Debtors-Appellants.

Mariah E. Murphy (Anthony C. Kaye and Steven D. Burt with her on the briefs),
Ballard Spahr LLP, Salt Lake City, Utah, for Appellee Citibank.

Kevin R. Anderson, Salt Lake City, Utah, for Trustee-Appellee.

Tara Twomey, San Jose, California, for Amicus Curiae National Association of
Consumer Bankruptcy Attorneys.


Before GORSUCH, HOLMES, and MATHESON, Circuit Judges.


GORSUCH, Circuit Judge.


      Like so many these days, Stephanie and Kenneth Woolsey owe more money

on their home than it’s worth. In fact, the value of their home doesn’t come close

to covering the balance due on their first mortgage, much less the amount they

owe on a second. And it’s that second mortgage, held by Citibank, at the center

of our case. After the Woolseys sought shelter in bankruptcy, they prepared a

Chapter 13 repayment plan. In their plan, they took the position that the

bankruptcy code voids Citibank’s lien because it is unsupported by any current

value in the home. Naturally, Citibank didn’t take well to the Woolseys’

intentions. The bank objected to the Woolseys’ plan and eventually persuaded the

bankruptcy court to reject it. Later the district court, too, sided with Citibank and

now the question has found its way to us.




                                        -2-
      Before us, though, the Woolseys don’t just shrink from, they repudiate the

only possible winning argument they may have had. They choose to pursue

instead and exclusively a line of attack long foreclosed by Supreme Court

precedent. To be sure, the Woolseys argue vigorously and with some support that

the Supreme Court has it wrong. But, as Justice Jackson reminds us, whether or

not the Supreme Court is infallible, it is final. See Brown v. Allen, 

,

540 (1953) (Jackson, J., concurring in the result). And it belongs to that Court,

not this one, to decide whether to revisit its precedent. For now, and like the

other judges to have passed on this case so far, we are obliged to apply the

Court’s current case law and that leads us, inexorably, to affirm.

                                        ***

      But before we can get to all that, there’s a jurisdictional snarl we have to

untangle first. After Citibank successfully objected to the Woolseys’ initial

repayment plan, the bankruptcy court issued an order rejecting it. That order, of

course, was hardly an appealable final decision spelling the end to things in

bankruptcy court: it promised only more litigation until an amended repayment

plan could win the bankruptcy court’s approval. See Simons v. FDIC (In re

Simons), 

, 645 (10th Cir. 1990). All the same, the Woolseys

appealed it to the district court. And this they could do because 28 U.S.C.

§ 158(a)(3) permits interlocutory appeals in these particular circumstances. For

its part, however, the district court soon issued a summary order affirming the

                                        -3-
bankruptcy court’s rejection of the Woolseys’ initial plan, and it is that decision

the Woolseys now seek to appeal to our court.

      And that raises this question: Do we have the power to hear an

interlocutory appeal of an interlocutory appeal? By what authority might we

entertain an appeal from the district court of an interlocutory order regarding a

matter pending in bankruptcy court? To be sure, the Woolseys could have sought

permission to proceed to this court under the general interlocutory appeal statute,

28 U.S.C. § 1292(b). See Conn. Nat’l Bank v. Germain, 

, 254

(1992). But they didn’t. Instead and at their behest, the district court purported

to certify its interlocutory appeal for a further interlocutory appeal to this court

under 28 U.S.C. § 158(d)(2)(A). Can a district court do that?

      When a case is properly certified by the bankruptcy court, district court or

bankruptcy appellate panel, Congress through § 158(d)(2)(A) has clearly given

this court the power to hear “appeals described in the first sentence of [§

158(a)].” The difficulty is that the “appeals described” in the first sentence of

§ 158(a) are not appeals from the district court, but appeals directly from the

bankruptcy court. So it’s evident enough that § 158(d)(2)(A) gives us the

authority to hear appeals straight from the bankruptcy court, leapfrogging over

the district court or bankruptcy appellate panel in order to speed up the resolution

of dispositive legal questions. See Weber v. U.S. Tr., 

, 157-58 (2d

Cir. 2007). What’s less certain is whether the statute also permits us to hear

                                          -4-
interlocutory appeals from the district court’s disposition of an interlocutory

appeal of a bankruptcy court order, in this respect covering much the same ground

as § 1292(b).

      Fortunately, it turns out we don’t have to decide that question in this case.

We don’t because, while this appeal was wending its way to us, the bankruptcy

court confirmed an amended repayment plan the Woolseys submitted after their

initial plan voiding Citibank’s lien was rejected. The confirmation of an amended

plan brought the bankruptcy proceedings to a close, surely constituting a final

order subject to appeal. See 28 U.S.C. § 158(d)(1). Indeed, in the world of

bankruptcy proceedings — a world where cases continue on in many ways for

many years and lack the usual final judgment of a criminal or traditional civil

matter — confirmation of an amended plan “is as close to the final order as any

the bankruptcy judge enters.” See Interwest Bus. Equip., Inc. v. U.S. Tr. (In re

Interwest Bus. Equip., Inc.), 

, 315 (10th Cir. 1994) (internal quotation

marks omitted).

      Neither does the fact the Woolseys filed their notice of appeal in this court

prematurely — after the district court decided its appeal but before the

bankruptcy court confirmed the Woolseys’ amended plan — deny them the right

to appeal. In the multi-layered appellate world of bankruptcy practice this

problem recurs not infrequently. And this circuit has responded by holding that,

at least absent any indication of potential prejudice, a premature notice of appeal

                                        -5-
involving a bankruptcy matter, even one (like this one) with an interstitial stop in

the district court, ripens and becomes effective once “a final order approving [a]

plan[] of reorganization” is entered. Interwest, 23 F.3d at 315; see also Adelman

v. Fourth Nat’l Bank & Tr. Co. (In re Durability, Inc.), 

, 265-66

(10th Cir. 1990). This court’s precedent, moreover, finds analogies of various

sorts in most other circuits. See, e.g., Community Bank, N.A. v. Riffle, 

, 173-74 (2d Cir. 2010) (per curiam); Rains v. Flinn (In re Rains), 

, 900-01 (9th Cir. 2005); Watson v. Boyajian (In re Watson), 

, 5

(1st Cir. 2005); Official Comm. of Unsecured Creditors v. Farmland Indus., Inc.

(In re Farmland Indus., Inc.), 

, 649-50 (8th Cir. 2005); In re Rimsat,

Ltd., 

, 1044 (7th Cir. 2000); In re Emerson Radio Corp., 

, 53 (3d Cir. 1995); 16 Charles Alan Wright, Arthur R. Miller & Edward H.

Cooper, Federal Practice and Procedure § 3926.2 at 290 (2d ed. 1996). After all,

the bankruptcy court’s refusal to confirm the Woolseys’ initial plan has now

become a final and irrevocable decision, no longer subject to reconsideration

there: the bankruptcy proceeding is closed. The district court has likewise

finished its work and can’t revise any decision we render. Neither have the

parties identified any prejudice anyone would suffer by taking up the appeal now.

Cumulatively, it’s clear everything that could be done below has been done.

      But even this analysis, we must acknowledge, isn’t without its wrinkles.

The notion that proceedings in a district court cumulatively might give rise to a

                                         -6-
final judgment was touched upon in FirsTier Mortgage Co. v. Investors Mortgage

Insurance Co., 

 (1991). There, the Court interpreted Fed. R. App. P.

4(a)(2)’s instruction that a “notice of appeal filed after the announcement of a

decision or order but before the entry of the judgment or order shall be treated as

filed after such entry and on the day thereof.” 498 U.S. at 272-73. While holding

that the Rule allowed the appeal at issue, the Court proceeded, in a statement that

may or may not have been essential to its holding, to say the Rule does not permit

a premature notice of appeal from a “clearly interlocutory decision — such as a

discovery ruling or a sanction order under Rule 11” because a “belief that such a

decision is a final judgment would not be reasonable.” Id. at 276 (emphasis in

original); see Gonzales v. Texaco Inc., 344 F. App’x 304, 307 (9th Cir. 2009)

(unpublished) (suggesting all this language is dicta). Instead, the Court said, the

Rule permits a premature notice of appeal only from decisions that themselves

“would be appealable if immediately followed by the entry of judgment” because

“[i]n these instances, a litigant’s confusion” about the finality of the case is

“understandable, and permitting the notice of appeal to become effective when

judgment is entered does not catch the appellee by surprise.” FirsTier Mortg.

Co., 498 U.S. at 276 (emphasis in original).

      Whether and to what degree this discussion, even if controlling, pertains to

bankruptcy practice is an open question. In the first place, it is a matter of some

debate whether Rule 4(a)(2) — and so FirsTier’s gloss on it — supplies the sole

                                         -7-
means for a court of appeals to secure jurisdiction over a prematurely filed

appeal. The Rule, some argue, is but a rule of practice, not a limit on our

statutory jurisdiction, and it might be supplemented by Fed. R. App. P. 2 when

necessary to allow courts to hear prematurely filed appeals that may not satisfy its

particular terms but involve sufficiently final decisions that we are statutorily

entitled to hear them. Compare Khan v. Attorney Gen., No. 11-1789, 

, at *3 n.2 (3d Cir. Aug. 14, 2012) (citing Lazy Oil Co. v. Witco Corp.,

, 587 (3d Cir 1999) so arguing), with Outlaw v. Airtech Air

Conditioning & Heating, 

, 160 n.2 (D.C. Cir. 2005) (disagreeing).

One might debate, as well, whether and to what degree the Rule needs to be

supplemented, by means of Rule 2 or otherwise, to fit the peculiarities of

bankruptcy practice. The Rule seems to assume the court “announc[ing] a

decision or order” is the same one that later enters “the judgment or order”

embodying that previous announcement. Yet here of course we have an appeal

from a decision announced by a district court concerning a final judgment (as it

were) later entered by the bankruptcy court, no unusual situation in bankruptcy

practice but surely not typical in the criminal or traditional civil practice to which

the Rule seems to speak. And one might wonder whether compliance with Rule

4(a)(2) can be forfeited or waived if it does not describe the outer limits of our

statutory jurisdiction. What it might say about any of this is surely a point for

debate, but it is a curiosity all the same that Interwest and most related cases in

                                         -8-
other circuits make no mention of the Rule (one way or the other) when

discussing appeals involving premature notices of appeal in the bankruptcy

context. And, perhaps like litigants in those cases, the litigants in ours have not

suggested the Rule or FirsTier imposes any impediment to entertaining this

appeal.

      Beyond even all this, however, FirsTier’s relevance in bankruptcy practice

is an open question for still a different reason: it isn’t clear whether a circuit

entertaining an appeal from a district court even needs to resort to the cumulative

finality doctrine — or for that matter § 1292 or § 158(d)(2)(A), two other

possibilities we’ve already discussed. When it comes to bankruptcy matters,

Congress in § 158(d)(1) gives circuit courts jurisdiction to hear “final decisions”

as well as final judgments, orders and decrees “entered under subsection[] (a).”

Turning to § 158(a), Congress there gives district courts authority to entertain

appeals from final judgments, orders and decrees, and certain interlocutory orders

and decrees from the bankruptcy court. Meanwhile, though, no mention is made

of “final decisions.” Given this arrangement, one might wonder whether this

court possesses jurisdiction under § 158(d)(1) to review a “final decision” of the

district court “entered under subsection[] (a),” with respect to a bankruptcy

court’s interlocutory order — all without the necessity of any final bankruptcy

court order. On this theory, the phrase “decision . . . entered under subsection[]

(a)” as it appears in § 158(d)(1) permits us to review a district court’s decision on

                                          -9-
a purely interlocutory issue. As applied to this case, the district court’s “final

decision” rejecting the Woolseys’ initial plan is enough to afford us jurisdiction:

the finality of the bankruptcy court’s proceedings is immaterial. See Wright,

Miller, & Cooper, supra, § 3926.2 at 279-80 (discussing merits and demerits of

this interpretation).

      The layering of appeal on appeal in the bankruptcy context, where our

usual ideas of finality are already put to the test, surely invites many and

interesting questions. But while we cannot ignore them, neither must we decide

them today. We don’t have to do so because, even assuming a final bankruptcy

proceeding is required to trigger our own jurisdiction under § 158(d)(1), and even

assuming that Rule 4(a)(2) is the only means by which we may entertain cases

involving a prematurely filed notice of appeal, this case satisfies the test this

court has set forth for satisfying the Rule.

      Though FirsTier’s cryptic and arguably tangential discussion about the

limits of Rule 4(a)(2) is open to many different understandings, in Hinton v. City

of Elwood, 

 (10th Cir. 1993), this court explained its understanding

that, under the Rule, “a premature notice of appeal retains its validity only when

the order appealed from is likely to remain unchanged in both its form and its

content.” Id. at 778. Admittedly, Hinton did not discuss FirsTier but it does

post-date FirsTier and so controls our analysis of the Rule. While a panel of this

court may sometimes recognize that a prior panel decision has been clearly

                                         - 10 -
overruled by an intervening Supreme Court decision, see, e.g., Currier v. Doran,

, 912 (10th Cir. 2001), we are aware of no existing Tenth Circuit

authority that might allow this panel to overrule another panel simply because the

earlier panel didn’t mention a still earlier and (possibly) relevant Supreme Court

decision.

      Hinton’s test, moreover, is clearly satisfied where, as here, the appealed

order from the district court (or Bankruptcy Appellate Panel) resolves the only

outstanding issue in what has, by the time it reaches us, become an otherwise

completed bankruptcy proceeding. In these circumstances, the challenged order

not only amounts to the order of a superior authority the bankruptcy court could

not undo later, it resolves the only appealed issue to arise out of the now

concluded bankruptcy proceedings and so could not be changed in any event. In

this situation, no one can claim serious surprise that the appealed order is

amenable to appellate review. Indeed, to hold otherwise might require the

appealing party to undertake the useless gesture of appealing again to the district

court the very same order it just issued, all so it could get the matter back to this

court, surely a paper pushing process no one would reasonably expect the law to

require. Because of the layering effect of review in bankruptcy, our situation is

very different from an attempt to appeal a district court’s discovery or Rule 11

sanctions order discussed in FirsTier that no party could reasonably think ended




                                        - 11 -
the case. No similar injustice or circumvention of traditional concepts of finality

are in play.

      Neither, in any event, could we hold otherwise. As we’ve already

explained, Interwest, like so many similar decisions in other circuits, expressly

authorizes bankruptcy appeal in the circumstances we face. And we are of course

bound by that precedent, as well. To be sure, and as we’ve already indicated,

Interwest (like Hinton) does not discuss FirsTier. But (again like Hinton)

Interwest post-dates FirsTier and we see no fair way by which we might avoid its

teachings.

      Though we are able, ultimately and after the application of some elbow

grease, to untie the jurisdictional knot and reach the merits of this case, we have

paused to describe along the way a number of unresolved and beguiling questions

so future litigants and district courts are not left unwarned or unwary — about the

questions surrounding § 158(d)(2)(A)’s reach, the curious discrepancies of

§§ 158(a) and (d), the anomalies of Rule 4(a)(2), and the still not entirely

explored frontiers of FirsTier.

                                        ***

      Now to the merits. May Chapter 13 debtors like the Woolseys void a state

law lien secured by no remaining value in the collateral? By all appearances, the

path to answering that question would seem to begin and end with the language of

11 U.S.C. § 506(d), a provision of the bankruptcy code allowing any debtor in

                                        - 12 -
bankruptcy, regardless under which specific Chapter the debtor proceeds, to

remove (or “strip off”) certain liens. Section 506(d) explains that, “[t]o the extent

that a lien secures a claim against the debtor that is not an allowed secured claim,

such lien is void.” Or, put differently, if Citibank’s claim isn’t an “allowed

secured claim,” the statute appears to allow the Woolseys to void the lien just as

they originally proposed.

      We are convinced Citibank’s claim is an “allowed” one. Section 502

provides that a claim filed by a creditor is “allowed” if nobody objects — and

even if there is an objection, a claim is still allowed unless stated otherwise in

§ 502. 11 U.S.C. § 502(a)–(b). The parties avidly dispute whether the Woolseys

ever properly objected to Citibank’s claim, but nothing turns on their fight. It

doesn’t because, even assuming the Woolseys did object, no provision in § 502

“disallows” the bank’s claim. Seeking to suggest otherwise, the Woolseys point

to § 502(b)(1). That provision “disallows” claims that are “unenforceable against

the debtor and property of the debtor, under any agreement or applicable law for a

reason other than because such claim is contingent or unmatured.” But it’s

beyond question that Citibank’s claim is a valid mortgage enforceable under Utah

law — and that leaves § 502(b)(1) without any bite here. Alternatively, the

Woolseys say Citibank’s claim is disallowed by § 506(a), if not by § 502.

Flipping forward a few pages to § 506(a), however, reveals that it has precisely

nothing to say about whether a claim is allowed or disallowed. By its terms

                                         - 13 -
§ 506(a) applies only to claims that are already “allowed,” making abundantly

clear it doesn’t “disallow” anything.

      Even accepting that Citibank’s claim is “allowed,” though, there remains

the question whether it is “secured.” And on that question § 506(a)(1) turns out

to have quite a lot to say:

      An allowed claim of a creditor secured by a lien on property in which the
      estate has an interest . . . is a secured claim to the extent of the value of
      such creditor’s interest in the estate’s interest in such property . . . and is an
      unsecured claim to the extent that the value of such creditor’s interest . . .
      is less than the amount of such allowed claim.

      The thrust of § 506(a) is to classify allowed claims (or portions of allowed

claims) as either secured or unsecured, which in turn affects how the bankruptcy

code treats them. The statute explains that for purposes of federal bankruptcy law

a “secured claim” requires something more than a security interest recognized by

state law. A claim, even if secured by a valid state law lien on property, qualifies

as “secured” for purposes § 506(a) and federal bankruptcy law only to the extent

it is supported by value in the collateral. United States v. Ron Pair Enters., Inc.,

, 239 (1989). To the extent that the lien is supported by some but

not enough value to cover the whole debt, § 506(a) splits the claim in two,

creating a secured claim and an unsecured claim. Id. at 239 n.3. So even if a lien

qualifies as a valid security interest under state law, it gives rise to a “secured

claim” for purposes of federal bankruptcy law only if and to the extent it is

supported by value in the underlying property.

                                         - 14 -
         All this would seem to bring us within a simple syllogism from the end of

this case. The code allows debtors to void liens that aren’t “allowed” and

“secured” claims. See 11 U.S.C. § 506(d). The code defines “allowed” claims.

See id. § 502. It also defines “secured” claims. See id. § 506(a). And so one

might be forgiven for thinking any lien either “disallowed” under § 502 or

“unsecured” under § 506(a) would be void under § 506(d). Because Citibank’s

junior lien isn’t backed by any value in the home, Citibank holds only an allowed

unsecured claim and so its lien would appear to be voidable, just as the Woolseys

argue.

         But the law in this corner of bankruptcy practice doesn’t follow such a

straight path. It doesn’t because of Dewsnup. See Dewsnup v. Timm, 

 (1992). There, the Supreme Court considered § 506(d) in the context of a

Chapter 7 bankruptcy proceeding and concluded that the term “allowed secured

claim” means a claim “allowed” under § 502 and “secured” by a lien enforceable

under state law. So it is, the Court held, value in the collateral has no bearing on

the lien-voiding language of § 506(d): any lien secured under state law must be

respected and protected from removal. Id. at 417. And it is precisely because of

Dewsnup’s holding that the bankruptcy court and district court refused the

Woolseys’ effort to remove Citibank’s lien using § 506(d).

         Now, one might well ask: How can it be that to qualify as “secured claim”

in § 506(a) some value is needed but a mere three subsections later in § 506(d),

                                         - 15 -
value is irrelevant to whether a claim is “secured”? It’s surely a topsy-turvy

result to give these two related provisions in the same statutory section entirely

different (even opposing) meanings. After all, it defies the Supreme Court’s own

“normal rule of statutory construction that identical words used in different parts

of the same act are [presumed] to have the same meaning.” Sullivan v. Stroop,

, 484 (1990) (internal quotation marks omitted).

      Even so, the Supreme Court found § 506(d)’s use of the term “secured”

sufficiently “ambiguous” that it felt at liberty to overlook these problems. And

the Court took the distinctly unusual step of finding the liberating ambiguity

based on no more than the fact the litigants before it happened to disagree over

the statute’s meaning — an ailment surely most afflicting every statutory

interpretation question in our adversarial legal system. Dewsnup, 502 U.S. at

416; see also id. at 422-23 (Scalia, J., dissenting) (criticizing the idea that a

statutory ambiguity might be “achieved by being the subject of disagreement

between self-interested litigants”). Still, in light of the ambiguity so generated,

the Court felt free to strike out to interpret § 506(d) on its own, unchained by

§ 506(a)’s plain language.

      Turning to historical practice for guidance, the Court said that state law

liens in Chapter 7 cases, at least before the enactment of the current bankruptcy

code, “pass[ed] through the bankruptcy case unaffected,” retaining their force

whether backed by value or not. Id. at 418. Without some indication in the

                                         - 16 -
legislative history that Congress intended to alter this practice, the Court decided

to interpret § 506(d) the same way in order to avoid “effect[ing] a major change

in pre-[c]ode practice.” Id. at 418. The Court also worried that any other result

would bestow upon the debtor the “windfall” of any increase in value during the

pendency of the bankruptcy, value more appropriately belonging to the lienholder.

Id. at 417.

      Even on their own terms these rationales are open to question. Whatever

pre-code practice looked like, it would seem to have (at best) limited interpretive

significance today, given that Chapter 7 indubitably permits liens to be removed

in many situations. See Harmon v. United States, 

, 581 (8th Cir.

1996) (collecting examples); In re Penrod, 

, 461-62 (7th Cir. 1995).

And it’s far from clear how much we have to worry about the debtor winning a

windfall: in most Chapter 7 cases it will be the remaining unsecured creditors

rather than the debtor who will reap any appreciation in the property’s value. See

Dewsnup, 502 U.S. at 422 n.1 (Scalia, J., dissenting); see also David Gray

Carlson, Bifurcation of Undersecured Claims in Bankruptcy, 70 Am. Bankr. L.J.

1, 10-11 (1996). Even more fundamentally still, when it comes to interpreting

statutes the Court itself has repeatedly instructed that pre-enactment practice is

relevant only “to the interpretation of an ambiguous text” and holds no sway

when the statutory language is clear. RadLAX Gateway Hotel, LLC v.




                                        - 17 -
Amalgamated Bank, 

, 2073 (2012). And the language of § 506(a)

and (d) does seem pretty plain.

       All this has led Justice Thomas to observe that Dewsnup has created more

than a little “methodological confusion,” confusion “enshroud[ing] both the

Courts of Appeals and, even more tellingly, Bankruptcy Courts, which must

interpret the Code on a daily basis.” Bank of Am. Nat’l Trust & Sav. Ass’n v. 203

North LaSalle Street P’ship, 

, 463 (1999) (Thomas, J., concurring in

the judgment). In fact, both Dewsnup’s decision to depart from the plain

language of § 506(a) and the rationales it supplied for doing so have engendered

many critics. 1

       But this much still is clear. Right or wrong, the Dewsnuppian departure

from the statute’s plain language is the law. It may have warped the bankruptcy

code’s seemingly straight path into a crooked one. It may not be infallible. But

until and unless the Court chooses to revisit it, it is final.

       1
         For a sampling of these criticisms, and beyond those found in the
powerful dissent in Dewsnup itself, see, e.g., Bank of Am. Nat’l Trust, 526 U.S. at
462-63 (Thomas, J., concurring in the judgment); Cunningham v. Homecomings
Fin. Network (In re Cunningham), 

, 245-46 (Bankr. D. Md. 2000);
Dever v. IRS (In re Dever), 

, 138 (Bankr. C.D. Cal. 1994); Lawrence
Ponoroff & F. Stephen Knippenberg, The Immovable Object Versus the
Irresistible Force: Rethinking the Relationship Between Secured Credit and
Bankruptcy Policy, 

. Rev. 2234 (1997); Carlson, supra, at 12-20; Barry
E. Adler, Creditor Rights After Johnson and Dewsnup, 10 Bankr. Dev. J. 1, 10-12
(1993); Mary Josephine Newborn, Undersecured Creditors in Bankruptcy:
Dewsnup, Nobelman, and the Decline of Priority, 25 Ariz. St. L.J. 547 (1993);
Margaret Howard, Dewsnupping the Bankruptcy Code, 1 J. Bankr. L. & Prac. 513
(1992).

                                          - 18 -
      That doesn’t stop the Woolseys from trying to argue otherwise. In their

view, Dewsnup controls the meaning of the term “secured” under § 506(d) only in

Chapter 7 cases. The very same term in § 506(d), they contend, should be given

an entirely different meaning when it comes to handling Chapter 13 cases —

requiring proof of value to avoid lien removal, just as the plain language of

§ 506(a) suggests. And we must admit their argument isn’t entirely without

appeal.

      Take Dewsnup’s reasoning first. When interpreting § 506(d), Dewsnup

relied heavily on perceptions about Chapter 7 practice. But, of course, § 506(d)

applies equally to bankruptcies like the Woolseys’ proceeding under Chapter 13.

And as the Woolseys point out, very different considerations are at work in

Chapter 7 liquidation bankruptcies than in reorganization bankruptcies under

Chapter 13. In a Chapter 7 bankruptcy, the debtor liquidates his non-exempt

assets to provide for immediate repayment of as much of his debt as possible, and

in exchange receives immediate relief from dischargeable debt. See Marrama v.

Citizens Bank of Mass., 

, 367 (2007). The satisfaction of secured

creditors comes primarily from a foreclosure sale of the collateral, and for this

reason under pre-code practice there wasn’t usually much reason to allow the

debtor to void a lien on property he will be forced to surrender in any event. By

contrast, in a Chapter 13 case the debtor’s obligations are not met primarily

through liquidation. Instead of selling his assets to meet his debts, the debtor has

                                        - 19 -
to commit a portion of his future income for a period up to five years as part of

the bankruptcy repayment plan. In exchange, “[t]he benefit to the debtor of

developing a plan of repayment under Chapter 13, rather than opting for

liquidation under Chapter 7, is that it permits the debtor to protect his assets.”

H.R. Rep. No. 95-595, at 118 (1977), reprinted in 1978 U.S.C.C.A.N. 5963, 6079.

      All this suggests that lien stripping has a very different role to play in

Chapter 13 than in Chapter 7. While voiding a lien may afford few benefits in a

Chapter 7 proceeding, it may be more integral to achieving Chapter 13’s goals.

After all, if the law always precluded lien stripping in the Chapter 13 context, a

debtor hoping to keep his property would have to provide for full repayment of a

lien no matter how large, even if his debt is secured by worthless collateral.

Faced with the prospect of paying much more than the property is worth under a

Chapter 13 plan, many more debtors would likely throw up their hands and simply

opt for liquidation. And this result would run contrary to Congress’s preference

for individual debtors to use Chapter 13 instead of Chapter 7. See id. at 118. It

would undercut Congress’s aim to allow individual debtors “to retain the pride

attendant” on satisfying a repayment plan rather than face the “stigma” of

liquidation. Id. It would thwart Congress’s stated purpose to change “the

treatment of secured creditors” in Chapter 13 to focus on “the true value of the

goods” secured as collateral rather than simply the lien’s “value as leverage”

against the debtor. Id. at 124. And it would not even obviously help creditors as

                                         - 20 -
a whole: because Chapter 13 debtors commit future income to debt repayment,

unsecured creditors’ losses are often “significantly less than [when] debtors opt

for” liquidation. Id. at 118; see also In re McDonald, 

, 614 (3d Cir.

2000). For all these reasons, and as Dewsnup itself recognized, pre-code practice

in reorganization cases like those under Chapter 13 (unlike pre-code Chapter 7

liquidation practice) often permitted liens to be stripped down to the value of the

collateral. 502 U.S. at 418-19. 2

      Not only does Dewsnup’s reasoning rest on peculiarities of the Chapter 7

context bearing little relevance to Chapter 13 practice, the case expressly instructs

us to read its holding narrowly. It tells us that its holding is limited to “the case

before us” and that “other facts . . . await their legal resolution on another day.”

Id. at 417. It adds candidly that the Court found it impossible to “interpret[] the

statute in a single opinion that would apply to all possible fact situations.” Id. at

416. And taking up this invitation to give the decision a crabbed reading, every

federal court of appeals to consider the question has already refused to extend


      2
         See also Haberman v. St. John Nat’l Bank (In re Haberman), 

, 1213 (10th Cir. 2008); Enewally v. Wash. Mut. Bank (In re Enewally), 

, 1170 (9th Cir. 2004) (“The rationales advanced in the Dewsnup
opinion for prohibiting lien stripping in Chapter 7 bankruptcies . . . have little
relevance in the context of rehabilitative bankruptcy proceedings under Chapter
11, 12, and 13, where lien stripping is expressly and broadly permitted”)
(quotation omitted); Wade v. Bradford, 

, 1128 (10th Cir. 1994);
Harmon, 101 F.3d at 581-82 & n.4; Margaret Howard, Secured Claims in
Bankruptcy: An Essay on Missing the Point, 23 Cap. U. L. Rev. 313, 314-16
(1994).

                                        - 21 -
Dewsnup’s definition of the term “secured claim” to other statutory provisions

using that term in Chapter 13, where the focus is on reorganization rather than

liquidation. 3 This same pattern — of circuits distancing themselves from

Dewsnup — recurs in Chapter 11 and Chapter 12 reorganization cases. 4 Most

notably, the Supreme Court itself has declined to extend Dewsnup’s

understanding of the term “secured claims” when it appears in Chapter 13. See

Nobelman v. Am. Sav. Bank, 

, 328 (1993) (“secured claim” in

§ 1322(b)(2) is defined with reference to § 506(a) and the value of the collateral);

Assocs. Commercial Corp. v. Rash, 

, 960 (1997) (same in

§ 1325(a)(5)(B)). So it is that Dewsnup has lost every away game it has played:

its definition of “secured claim” has been rejected time after time elsewhere in the

code and seems to hold sway only in § 506(d).

      Building on all this, the Woolseys invite us to hand Dewsnup a loss even on

its home court, within § 506(d) itself. Dewsnup suggests that the term “secured

claim” may mean something different in § 506(d) than it does in the rest of the

bankruptcy code, and many circuits and the Supreme Court itself have now



      3
         See, e.g., Haberman, 516 F.3d at 1213; Enewally, 368 F.3d at 1169-70;
Bartee v. Tara Colony Homeowners Assoc. (In re Bartee), 

, 291 n.21
(5th Cir. 2000).
      4
        Chapter 11: Wade v. Bradford, 

, 1128-30 (10th Cir. 1994);
In re Heritage Highgate, Inc., 

, 144 (3d Cir. 2012). Chapter 12:
Okla. ex rel. Comm’rs of the Land Office v. Crook (In re Crook), 

,
539 n.1 (10th Cir. 1992); Harmon, 101 F.3d at 582.

                                       - 22 -
(repeatedly) held as much. All that, of course, is curious enough in light of the

“normal rule” that identical words bear identical meaning throughout a statutory

structure. Sullivan, 496 U.S. at 484. But, the Woolseys say, we should go now a

step further and read the term “secured claim” to mean two different things even

within § 506(d) itself. When a Chapter 7 case comes along, they say, the term

should mean what Dewsnup says it means: any claim secured under state

property law is protected from removal, even if backed by no value. But when a

court is faced with a Chapter 13 case, they argue, the term should be read to

require proof of value before a lien is protected from removal in accord with

§ 506(a). They stress that Dewsnup’s rationales are limited to Chapter 7 cases

and that Dewsnup itself seems to suggest the term “secured claims” may be

interpreted to mean different things in different “fact situations.” 502 U.S. at

416.

       Though we must admit the Woolseys’ invitation to undo Dewsnup in this

way has its attractions, it’s an especially odd invitation to issue, let alone for this

court to accept. After all, it violates yet another and even more elementary rule

of statutory interpretation: the rule against “[a]scribing various meanings to a

single iteration” of a statutory term in different applications. Ratzlaf v. United

States, 

, 143 (1994) (emphasis added) (internal quotation marks

omitted). Though giving a term different meanings in different but related

statutes is one thing and disfavored enough, in recent years the Supreme Court

                                          - 23 -
has suggested that giving a single use of a term different meanings is another

thing altogether, a ploy not just frowned upon but methodologically incoherent

and categorically prohibited: “To give these same words a different meaning for

each category [of cases to which they apply] would be to invent a statute rather

than interpret one.” Clark v. Martinez, 

, 378 (2005); see also id. at

386 (rejecting “the dangerous principle that judges can give the same statutory

text different meanings in different cases”); United States v. Santos, 

, 522-23 (2008) (opinion of Scalia, J.) (Clark “forcefully rejected” the notion

that courts could “giv[e] the same word, in the same statutory provision, different

meanings in different factual contexts” (emphasis omitted)).

      Applying this rule, the Supreme Court has refused to give different

meanings to a single statutory term even when the case for doing so is far

stronger than the case the Woolseys are able to muster here. In Clark, for

example, the Supreme Court held that when a statutory provision is given a

limiting construction to avoid a serious constitutional question arising from one

of its potential applications, that interpretation governs all applications of the

provision — even those that do not raise the same constitutional concerns. Clark,

543 U.S. at 377-78. While noting that “the statutory purpose and the

constitutional concerns” motivating the prior limiting construction were not

present in the case before it, the Court held this still “cannot justify giving the

same . . . provision a different meaning” in different factual circumstances. Id. at

                                         - 24 -
380 (emphasis in original). Rather, “the lowest common denominator, as it were,

must govern.” Id. Similarly, the Court has held, when a statute possesses both

criminal and civil applications a narrowing interpretation in a criminal case

driven by the rule of lenity must apply equally to civil litigants to whom lenity

would not ordinarily extend. Id. at 380 (citing United States v. Thompson/Center

Arms Co., 

, 517-518 & n. 10 (1992) (plurality opinion) and id. at 519

(Scalia, J., concurring in judgment)). If the rules of lenity and constitutional

avoidance — powerful and long-standing interpretive traditions — fail to justify

giving multiple meanings to a statutory term, it’s difficult to see how anything the

Woolseys offer us in this case might.

      Not only is the rule against multiple interpretations of the same statute well

entrenched, it is of special importance. Without it, even a statutory term used but

a single time in a single statute risks never settling on a fixed meaning. And this

surely would leave citizens at sea, only and always guessing at what the law

might be held to mean in the unique “fact situation” of the next case — a result in

no little tension with the rule of law itself. See id. at 382 (to accept that the same

statutory provision could have multiple meanings “would render every statute a

chameleon, its meaning subject to change . . . in each individual case”).

      Given all this, it’s perhaps no surprise that of all the circuit courts

approving of lien stripping in reorganization cases, not a single one has taken up

the Woolseys’ invitation to do so using § 506(d). Instead, they have relied

                                         - 25 -
exclusively on other statutory provisions particular to those chapters. See, e.g.,

Wade, 39 F.3d at 1129 (permitting lien stripping in Chapter 11 cases under 11

U.S.C. § 1129(b)(2)); Harmon, 101 F.3d at 583 (permitting lien stripping in

Chapter 12 cases under 11 U.S.C. § 1225(a)(5)); In re Lane, 

, 665

(6th Cir. 2002) (permitting lien stripping in Chapter 13 cases under 11 U.S.C.

§ 1322(b)(2)); see also 4 Collier on Bankruptcy, ¶ 506.06[1] (16th ed. 2011)

(“[T]here is no principled way to conclude that, although section 506(d) does not

authorize lien stripping in chapter 7 cases, it has a different meaning in chapter

11, 12, and 13 matters.”).

      In light of recent and unambiguous Supreme Court precedent repudiating

the interpretive move the Woolseys invite us to take, and because the Woolseys

never even pause to confront this precedent, we decline to follow where they wish

to lead. We do not doubt a strong argument can be made that the language and

logic of § 506 permit the Woolseys to void not only Citibank’s lien but any lien to

the extent it is unsupported by value in the collateral. But we fail to see any

principled way we might, as lower court judges, get there from here. Dewsnup

may be a gnarled bramble blocking what should be an open path. But it is one

only the Supreme Court and Congress have the power to clear away.

                                        ***

      Having said that much, we candidly acknowledge we thought at first there

might possibly be a way around Dewsnup, if not a way to plow through it so

                                        - 26 -
directly as the Woolseys urge. As we’ve already alluded to, many courts seeking

to avoid Dewsnup’s pinch have invoked provisions specific to the reorganization

chapters to permit the removal or stripping down of liens unsupported by value.

And when it comes to Chapter 13, many courts have already identified one

apparently promising candidate in § 1322(b)(2).

      Section 1322 sets forth the provisions a Chapter 13 reorganization plan may

contain (as well as some that it must). Relevant for our purposes is § 1322(b)(2),

which provides that the plan may

      modify the rights of holders of secured claims, other than a claim secured
      only by a security interest in real property that is the debtor’s principal
      residence, or of holders of unsecured claims, or leave unaffected the rights
      of holders of any class of claims[.]

      In broad strokes, this provision permits a debtor’s plan to modify the rights

of secured or unsecured creditors — all as part of Chapter 13’s general effort to

find a realistically achievable repayment plan. Of course, the power to modify

creditors’ claims is qualified — qualified both by other parts of § 1322 not at

issue here, and by § 1322(b)(2)’s own internal limitation on the modification of

secured claims in real property. But those exceptions aside, the first and third

clauses of § 1322(b)(2) arm the debtor with considerable power to modify the

rights of secured and unsecured creditors alike.

      At first we wondered whether this modification power might include the

ability to strip off a lien unsupported by value in its collateral. To be sure,


                                        - 27 -
§ 1322(b)(2) itself prohibits modification of the rights of the “holder of a secured

claim” supported by a lien on the debtor’s home. But the Supreme Court’s

decision in Nobelman v. American Savings Bank, 

 (1993), can be

read to suggest that for a claim to be “secured” and therefore trigger the anti-

modification clause in § 1322(b)(2), it must be supported by at least some value

in the collateral — just as § 506(a) (but not Dewsnup) says. Nobelman, 508 U.S.

at 328-31. Indeed, no fewer than six circuits have already read Nobelman this

way and held a debtor may invoke § 1322(b)(2) to remove a wholly unsecured

lien, even if that lien is secured against the debtor’s principal residence. See Lane

v. W. Interstate Bancorp (In re Lane), 280 F.3d at 665; Zimmer v. PSB Lending

Corp. (In re Zimmer), 

, 1221 (9th Cir. 2002); Pond v. Farm

Specialist Realty (In re Pond), 

, 127 (2d Cir. 2001); McDonald, 205

F.3d at 615; Bartee, 212 F.3d at 280; Tanner v. FirstPlus Fin., Inc. (In re

Tanner), 

, 1360 (11th Cir. 2000); see also 8 Collier on Bankruptcy,

supra, ¶ 1322.06[1][a][i].

      The hitch is, the Woolseys didn’t choose to pursue this line of argument

before this court in their initial briefs. What’s more, they have now expressly

repudiated it in a supplemental submission. This even though the bankruptcy

court offered a favorable discussion of lien stripping under § 1322(b)(2), uniform

circuit precedent endorses it, and an amicus brief in this case from the National

Association of Consumer Bankruptcy Attorneys ably argued the point. For their

                                        - 28 -
part, however, the Woolseys only glancingly referred to § 1322(b)(2) in their

opening briefs. Wanting to know more about that provision in light of all the

authority discussing it as a possible avenue for relief for debtors like the

Woolseys, we decided to ask the parties for supplemental briefing on the question

whether § 1322(b)(2) permits a Chapter 13 debtor to remove a wholly unsecured

lien even if § 506(d) does not. In response, the Woolseys made plain that they

wanted no part of the argument. They emphatically announced “[t]here is no

Code provision other than 11 U.S.C. §506(d) that declares void a wholly

unsecured lien.” Appellants’ Supp. Br. at 4.

      And that leaves us in an awkward place. There’s a potentially promising

argument for the Woolseys, one suggested by their own amicus, but it is one they

want no part of. Whatever our power to tackle the § 1322(b)(2) question in these

circumstances, nothing requires us to do so, to foist on litigants arms they so

avidly refuse to take up in the adversarial arena. So in deference to their wishes,

we opt today against forcing a § 1322(b)(2) argument onto the unwilling

Woolseys and leave that statute and meaning for another day when a bankruptcy

petitioner actually wants to pursue the question. In this case, we limit ourselves

to the question the Woolseys do want us to address: did the bankruptcy and

district courts err in holding that § 506(d) precluded them from removing

Citibank’s lien? The answer to that narrow question, we have seen, has to be no




                                         - 29 -
so long as Dewsnup remains the law. For that reason (and that reason alone) we

affirm. 5




       5
         Because this is a sufficient basis to affirm the bankruptcy court’s refusal
to confirm the plan proposing to remove Citibank’s lien, we have no need to
consider the bankruptcy court’s separate holding that the Woolseys’ plan was also
deficient for failing to include the so-called “lien retention” language in
§ 1325(a)(5)(B)(i)(II).

                                       - 30 -