concurring.
I agree that we are bound by our panel decision in NTP, Inc. v. Research in Motion, Ltd., in which we held that “direct infringement under section 271(a) is a necessary predicate for government liability under section 1498.” 418 F.3d 1282, 1316 (Fed.Cir.2005) (citing Motorola, Inc. v. United States, 729 F.2d 765, 768 n. 3 (Fed.Cir.1984)). However, the NTP proposition is, in my view, the result of an unchecked propagation of error in our case law, and its viability may eventually be challenged. Consequently, I write separately to emphasize that our decision today does not depend for its validity on NTP, as it is also supported by an independent line of reasoning.
I.
Before discussing this additional basis for today’s decision, I shall briefly address the ignoble history of the NTP proposition. An examination of the case law reveals that the proposition has its roots in Decca v. United States, in which we held that the United States could be liable under 28 *1354U.S.C. § 1498 only for direct infringement and not for indirect infringement:
Section 1498 expressly waives the Government’s sovereign immunity only with respect to governmental direct infringement of a patent. Nowhere in the section is active inducement of infringement or contributory infringement mentioned, either directly or by cross-reference to 35 U.S.C. §§ 271(b) and (c). A waiver of sovereign immunity must be strictly construed. Stated differently, the Government is not to be regarded as having waived its sovereign immunity by implication.
225 Ct.Cl. 326, 340, 640 F.2d 1156 (Ct.Cl. 1980). Thus, after Decca, a patentee who sought to establish a § 1498(a) claim had to prove “direct” infringement by the United States. I have no problem with this formulation of the rule, as it was the product of the Decca court’s thoughtful analysis.
It is important to note, however, the relationship between the phrases “direct infringement,” as used in Decca, id., and the phrase “direct infringement under § 271(a)” as used in NTP, 418 F.3d at 1316. That relationship is one of set to subset: infringement under 35 U.S.C. § 271(a) is a specific kind of “direct infringement.” Thus, while “direct infringement” is one of the requirements of an action against a private party under § 271(a), it is not the only requirement, with § 271(a) also requiring, inter alia, that the infringing activity occur “within the United States.”
Decca, of course, did not require that a § 1498(a) plaintiff satisfy all of the requirements of § 271(a). Indeed, Decca did not even mention § 271(a). Instead, it required only that the accused use constitute a direct, rather than an indirect, infringement. NTP, in contrast, states as a blanket proposition that all of the requirements of 35 U.S.C. § 271(a) must be satisfied in order to prove a 28 U.S.C. § 1498 claim against the government. See NTP, 418 F.3d at 1316.
How did such a sweeping generalization evolve from Decca? The answer, unfortunately, is that its evolution owes to this court’s imprecise recharacterization of the Decca holding in our decision in Motorola. See 729 F.2d 765, 768 n. 3. In Motorola, this court cited Decca at the end of a footnote as one of a number of cases supporting the very general proposition that the patent statutes are often inapplicable to actions arising under section 1498. Id. It stated that “the Government can only be sued for any direct infringement of a patent (35 U.S.C. § 271(a)), and not for inducing infringement by another (section 271(b)) or for contributory infringement (section 271(c)).” Id. (emphasis added) (citing Decca, 640 F.2d at 1167).
By equating the specific statutory provision “35 U.S.C. § 271(a)” with the more general concept of “direct infringement,” the Motorola court carelessly extended the Decca holding, thus laying the groundwork for the NTP proposition. There is distinct irony in the Motorola court’s being responsible for the creation of an inadvertent link between 35 U.S.C. § 271(a) and 28 U.S.C. § 1498(a), given that the entire point of the Motorola footnote was to support the proposition that it is erroneous to simply assume the existence of relationships between these two statutory sections. See id. at 768 (explaining that “ [although a section 1498 action may be similar to a Title 35 action, it is nonetheless only parallel and not identical”).
Although in Motorola the relationship between sections 271(a) and 1498(a) consti*1355tuted mere dicta, in NTP the relationship, as slightly recast, became the binding precedent of this court. In my view, such a sweeping rule of binding precedent ought not to be created by mere inadvertence. In turn, the resolution of the important issue before this court today—the scope of the government’s waiver of sovereign immunity as embodied in § 1498— ought not to be based solely on the basis of NTP.
In his concurring opinion, Judge Dyk attempts to find support for the NTP proposition. Citing certain legislative materials which state that the sole purpose of enacting § 1498(a) was to create jurisdiction (rather than additional liability), he argues that § 1498(a), even when viewed apart from, 1498(c), cannot result in “the government’s liability [being] broader than that of private parties.” Id. In other words, according to his concurrence the limits of government liability under § 1498(a) are circumscribed by the hypothetical liability of a private actor under § 271(a), and we can entirely ignore § 1498(c) in the analysis.
His reliance on legislative history is misplaced, however, given that the plain language of the text demands a contrary and unambiguous interpretation. As is explained below, the plain language of the statute reveals that 28 U.S.C. § 1498 is not a mere waiver of sovereign immunity with respect to the previously-existing remedy, which was provided by 35 U.S.C. § 281 for transgressions denominated by § 271 as “infringing” the exclusionary rights granted by § 154(a).
There can be little doubt that § 1498 is more than a mere waiver of sovereign immunity. The plain language of the statute speaks of the provision of a “remedy,” and the fourth paragraph of subsection (a) indicates that “remedy” means “right of action.” See id. (“[This section shall not convey a right of action on any patentee ... with respect to any invention discovered ... while in the employment ... of the United States ____”]) (emphasis added). Section 1498 is by no means a mere jurisdictional statute, and to the extent that materials in the legislative history may appear to suggest otherwise, they serve principally to illustrate why courts should often view legislative history with skepticism and rely on it only when the statute itself is ambiguous. If the NTP proposition can be sustained, it cannot be sustained on this ground.
Granted, in creating a new remedy, § 1498(a) inherently creates new rights. However, there is nothing about the text of § 1498(a) that either suggests or requires the existence of any substantive or conceptual link between these rights and remedies and those provided by the patent laws (pursuant to 35 U.S.C. §§ 154(a) and 281, respectively). The independence of these two sets of rights and remedies results from § 1498(a) conditioning the availability of a remedy (and the inherent creation of a corresponding right) only on the “invention [being] described in and covered by a patent” (emphasis added) and the invention’s being “used or manufactured by or for the United States without license of the owner thereof or lawful right .... ” (emphasis added). The “coverage” provided by a patent is, of course, defined by the scope of its claims. See, e.g., Johnson & Johnston Assocs. v. R.E. Serv. Co., 285 F.3d 1046, 1052 (Fed.Cir.2002) (expounding on the “fundamental principle that claims define the scope of patent protection”); see also SRI Int’l v. Matsushita Elec. Corp., 775 F.2d 1107, 1121 (Fed.Cir. 1985) (“It is the claims that measure the invention.” (citing Aro Mfg. Co., Inc. v. *1356Convertible Top Replacement Co., Inc., 365 U.S. 336, 339, 81 S.Ct. 599, 5 L.Ed.2d 592 (1961))). Accordingly, a § 1498 right and remedy exist whenever an accused embodiment, used or manufactured by or for the government, falls within the scope of the patent claims as properly construed.1 The “coverage” inquiry is independent of the nature and identity of the Title 35 rights that might have been available against a private infringer under the same circumstances.
Indeed, prior to the passage of § 1498(c), the remedy created by § 1498(a) was broader than that available against private parties in some instances. Again, the plain text of subsection (a) makes this clear. Whereas § 271(a) contains a territoriality restriction, § 1498(a) contains none. Indeed, if § 1498(a) had not created a cause of action against the government that was broader, with respect to overseas infringement, than that available against a private infringer, then it is difficult to understand the purpose of § 1498(c)—a subsection specifically barring § 1498 claims that “aris[e] in a foreign country.”
In summary, patentees have the following rights and remedies: (1) a 35 U.S.C. § 281/271 remedy (for which neither 28 U.S.C. § 1498(a) nor any other statute provides a waiver of sovereign immunity) for transgression of rights granted by 35 U.S.C. § 154, and (2) a 28 U.S.C. § 1498 remedy (assertable against the federal government) for transgressions of an inherently-defined right to be compensated for government manufacture or use of inventions covered by patents. Because § 1498(a) created new and independent rights and remedies, rather than merely waiving sovereign immunity with respect to an old remedy (i.e. § 281), one cannot argue that § 1498(a) a priori requires § 271(a) as an underlying and predicate cause of action.
In fact, not even NTP posited a relationship specifically between § 271(a) and § 1498, subsection (a). Rather the articulated relationship—the so-called “NTP proposition”—is between § 271(a) and § 1498 as a whole. See NTP, 418 F.3d at 1316 (holding that “direct infringement under section 271(a) is a necessary predicate for government liability under section 1498”). Not even NTP requires us to ignore the importance of § 1498, subsection (c), as Judge Dyk urges us to do in his concurrence.
I should explain that the per curiam opinion—which I join—does indeed state that the government’s “liability does not exist pursuant to § 1498(a)” and that “1498(a) bars Zoltek’s claims.” (emphasis added). However, this usage simply reflects the fact that subsection (a) creates the cause of action. Consequently, liability—if not precluded by subsection (c)— must arise pursuant to subsection (a). The court’s opinion today in no way endorses the view that subsection (c) is irrelevant, although we clearly do acknowledge the precedential effect of the NTP proposition, as articulated by the NTP court.
*1357Judge Dyk’s concurrence advances a second argument in support of the NTP proposition, namely that § 1498(a) was intended to be the “exact equivalent” of § 271(a), citing for support Richmond Screw Anchor Co. v. United States, 275 U.S. 331, 343, 48 S.Ct. 194, 72 L.Ed. 303 (1927). A careful reading of that case reveals, however, that the Court made no such pronouncement. Richmond Screw was a case about the assignability of patent claims. The backdrop was Rev. Stat. § 3477, which forbade inter alia the assignment of patent infringement claims against the United States that had not been reduced to judgment. Thus, under § 1498(a), as it existed prior to amendment in 1918, when a government contractor infringed a patent, the patent owner had both an assignable right of action against the contractor under § 271(a) and a claim against the government pursuant to § 1498(a), with the latter being nonassignable pursuant to Rev. Stat. § 3477.
The 1918 amendments to § 1498 eliminated the remedy against the contractor pursuant to § 271(a), confining the paten-tee to a suit against the government for his entire compensation. See Pub.L. No. 65-182, 40 Stat. 704, 705 (1918). The salient question was what would become of assigned claims against the contractor. Did Rev. Stat. § 3477, which barred the assignment of claims against the United States, prevent the assignee of a § 271(a) claim against a government contractor from pursuing a 1498(a) claim against the government pursuant to the 1918 Amendment? Troubled by the possibility of the 1918 Act effecting an uncompensated taking under the Fifth Amendment, the Court held that § 3477 did not apply “to the assignment of a claim against the United States which is created by the Act of 1918 in so far as the Act deprives the owner of the patent of a remedy against the infringing private contractor for infringements thereof and makes the Government indem-nitor for its ... infringements.” Richmond Screw, 275 U.S. at 346, 48 S.Ct. 194.
Contrary to Judge Dyk’s reading, the Court did not assert a blanket proposition that § 1498(a) must be viewed as being coextensive in all respects with § 271(a). It is true that the Court said, as quoted in his concurrence, that “[w]e must presume that Congress in the passage of the Act of 1918 intended to secure to the owner of the patent the exact equivalent of what it was taking away from him.” (emphasis added). The concurring opinion appears to assume—incorrectly—that “what it is taking away” refers to § 271(a) as a whole, leading Judge Dyk to conclude that the Act of 1918 (constituting § 1498, as amended) is “the exact equivalent” of § 271(a). However, the Court tells us what it was referring to by the words “what it was taking away,” and it was not referring to § 271(a) rights globally. Rather, it was referring—narrowly—to the assignability of rights pursuant to § 271(a). The next two sentences of the quoted passage, which the concurrence omits, make this point clear.
We must presume that Congress in the passage of the Act of 1918 intended to secure to the owner of the patent the exact equivalent of what it was taking away from him. It was taking away his assignable claims against the contractor for the latter’s infringement of his patent. The assignability of such claims was an important element in their value and a matter to be taken into account in providing for their just equivalent.
Id. at 345, 48 S.Ct. 194 (emphasis added). In its careful—and restrictive—analysis the Supreme Court actually declined the *1358opportunity to articulate a broad, sweeping relation between § 271(a) and § 1498.
The NTP court should not have articulated such a relationship, especially not through reliance on logically flawed dicta, and I can see neither the need nor the clear basis for us—at least in this case—to attempt to support through logic, post hoc, what the NTP court has wrought through folly. Perhaps the NTP proposition will in the future gain recognition as a useful expedient against the backdrop of a more thoroughly interpreted § 1498. For now, however, I view the proposition as being no more than an inadvertent leap of faith from the specific to the general. Accordingly, while I join the court’s opinion— recognizing that NTP binds us—my confidence that we are reaching the just and logical outcome stems not from it, but from the following analysis.
II.
I believe that this case poses two substantial questions of statutory interpretation. First, when is a patented method “used” under § 1498(a)? Second, does an infringement action “arise[] in a foreign country” under § 1498(c) if some steps of a claimed method are practiced abroad?
I conclude that a party “uses” a method claim within § 1498(a) only when it practices every step of the claimed method. Furthermore, an action for an infringing use of a patented method “arises in a foreign country” under § 1498(c) whenever any claimed step is practiced abroad. I would therefore have affirmed the trial court’s conclusion that § 1498 bars Zol-tek’s claims, even if we were not compelled to recognize NTP as precedent.
Section 1498 provides, in relevant part:
Whenever an invention described in and covered by a patent of the United States is used ... by or for the United States without license of the owner thereof or lawful right to use or manufacture the same, the owner’s remedy shall be by action against the United States in the United States Court of Federal Claims for the recovery of his reasonable and entire compensation for such use and manufacture____ For the purposes of this section, the use ... of an invention described in and covered by a patent of the United States by a contractor, a subcontractor, or any person, firm, or corporation for the Government and with the authorization or consent of the Government, shall be construed as use ... for the United States.
28 U.S.C. § 1498(a) (2000) (emphases added). At § 1498(c), Congress further limited the scope of claims under § 1498(a) with a foreign country exception. That limitation provides:
The provisions of this section shall not apply to any claim arising in a foreign country.
28 U.S.C. § 1498(c) (2000).
Both sections require interpretation. First, the court must determine the meaning of a “use,” as applied to a method claim, in § 1498(a). Second, the court must determine when a claim “arises” in a foreign country under § 1498(c). The trial court complicated this analysis by collapsing the two steps together.2 Although the two sections must be construed in harmony to preserve the integrity of the statutory scheme, in the first instance the questions of “use” and “arising” are separable. I address each issue in turn.
*1359A.
The court’s statutory analysis must give effect to Congress’s intent when enacting the statute. See Nat’l Labor Relations Bd. v. Lion Oil Co., 352 U.S. 282, 297, 77 S.Ct. 330, 1 L.Ed.2d 331 (1957) (observing that the court bears “a judicial responsibility to find that interpretation which can most fairly be said to be embedded in the statute, in the sense of being most harmonious with its scheme and with the general purposes that Congress manifested”); Doyon, Ltd. v. United States, 214 F.3d 1309, 1314 (Fed.Cir.2000). The analysis begins with the text of the statute. Robinson v. Shell Oil Co., 519 U.S. 337, 340, 117 S.Ct. 843, 136 L.Ed.2d 808 (1997); White v. Dep’t of Justice, 328 F.3d 1361, 1374 (Fed.Cir.2003). To fully understand the meaning of a statute, however, the court looks “not only to the particular statutory language, but to the design of the statute as a whole and to its object and policy.” Crandon v. United States, 494 U.S. 152, 158, 110 S.Ct. 997, 108 L.Ed.2d 132 (1990); accord Dunn v. Commodity Futures Trading Comm’n, 519 U.S. 465, 478, 117 S.Ct. 913,137 L.Ed.2d 93 (1997).
In this case the text of § 1498(a) is silent on the test for “use” of a patented method. The specific problem is a question of degree. In particular, one must decide whether “use” means practicing some steps of a claimed method, or instead requires practicing every claimed step.
At the outset I note that it is well-settled, as a matter of patent law, that an infringing use of a patented method requires practicing every step claimed in the method.3 See Smith & Nephew, Inc. v. Ethicon, Inc., 276 F.3d 1304, 1311 (Fed.Cir.2001) (“Infringement arises when all of the steps of a claimed method are performed[.]”); see also NTP, 418 F.3d at 1318 (“[A] process cannot be used ‘within’ the United States as required by section 271(a) unless each of the steps is performed within this country.”);4 accord Deepsouth Packing Co., Inc. v. Laitram Corp., 406 U.S. 518, 528, 92 S.Ct. 1700, 32 L.Ed.2d 273 (1972) (“[A] combination patent [asserted under § 271(a) ] protects only against the operable assembly of the whole and not the manufacture of its parts.”). The rule applies the long-settled principle, more commonly phrased in terms of products and the elements of device claims, that infringement requires practicing each and every limitation of the asserted patent claim. Sec V-Formation, Inc. v. Benetton Group SpA, 401 F.3d 1307, 1312 (Fed.Cir.2005) (recognizing the black-letter law that an accused product infringes a patent claim only if each and every limitation in the claim appears in the accused product); Becton Dickinson & Co. v. C.R. Bard, Inc., 922 F.2d 792, 796 (Fed.Cir.1990) (“To establish infringement of a patent, every limitation set forth in a claim must be found in an accused product or process exactly or by a substantial equivalent.”). “[A]ll [specified elements in a patent claim] must be regarded as material, leaving open only the question whether an omitted part is supplied by an equivalent *1360device or instrumentality.” Hubbell v. United States, 179 U.S. 77, 82, 21 S.Ct. 24, 45 L.Ed. 95 (1900). Nothing in the text of § 1498 warrants a contrary interpretation of an infringing use here. Indeed, the result is consistent with the narrow reading that a court must assign § 1498.
Because § 1498(a) waives sovereign immunity for infringement actions against the government, a court must give the statute a strict construction. See Lane v. Pena, 518 U.S. 187, 192, 116 S.Ct. 2092, 135 L.Ed.2d 486 (1996); United States v. Sherwood, 312 U.S. 584, 590, 61 S.Ct. 767, 85 L.Ed. 1058 (1941) (citing Schillinger v. United States, 155 U.S. 163, 30 Ct.Cl. 480, 15 S.Ct. 85, 39 L.Ed. 108 (1894)); Shoshone Indian Tribe v. United States, 364 F.3d 1339, 1346 (Fed.Cir.2004); RHI Holdings, Inc. v. United States, 142 F.3d 1459, 1461 (Fed.Cir.1998). The waiver “must be unequivocally expressed in statutory text” and the courts will not imply exceptions. Lane, 518 U.S. at 192, 116 S.Ct. 2092 (citing United States v. Nordic Village, Inc., 503 U.S. 30, 33-34, 37, 112 S.Ct. 1011, 117 L.Ed.2d 181 (1992) and Irwin v. Dep’t of Veterans Affairs, 498 U.S. 89, 95, 111 S.Ct. 453, 112 L.Ed.2d 435 (1990)). Consistent with these principles a court must resolve any ambiguity in the statute “in favor of immunity.” See United States v. Williams, 514 U.S. 527, 531, 115 S.Ct. 1611, 131 L.Ed.2d 608 (1995). Applying these principles to § 1498(a) makes the narrowest interpretation of an infringing “use” presumptively correct. Because I must narrowly construe the breadth of the statutory language, the sovereign immunity analysis confirms that the “use” of a claimed method under § 1498(a) requires practicing every step in the claim.
B.
Assuming Zoltek alleges an infringing use, the question becomes where the § 1498(a) claim arises. Under § 1498(c), Congress curtailed its sovereign immunity waiver to preclude § 1498(a) claims “arising in a foreign country.” The question here is the extent of this curtailment when the government practices a patented process having multiple steps.
In addressing this question, it is useful to imagine four factual scenarios. The first and easiest scenario is where all of the steps of the process were performed in this country—a situation in which there is no question that a claim, if any, must arise in the United States. The second scenario is where all of the steps were performed abroad, as is true in this case with the alleged infringement of the Nicalon fiber claims. (I address these claims in the next section.) Later in the opinion, I will address the meaning of 1498(c) in the context of the third and fourth scenarios, in which some, but not all steps were performed abroad; in the third scenario, the last step of the process is practiced abroad, and in the fourth scenario (as with the Tyranno fiber claims), the last step is practiced in the United States.
1.
I shall start with the Nicalon fiber claims, in which all of the infringing steps were performed abroad. Clearly the infringing activity is entirely outside the United States from a geographical perspective. Therefore, it does not occur “within” the United States as required by § 271(a), as the law has long recognized that patent infringement claims arise where the infringing activity is consummated. See, e.g., Deepsouth, 406 U.S. at 527, 92 S.Ct. 1700 (holding that final assembly abroad of a patented device cannot be an infringing manufacture within the meaning of § 271(a)). But this is not a *1361§ 271(a) action, and there is therefore no a priori reason why this court must interpret the “arising in a foreign country” language of § 1498(c) as being antonymous with the phrase “within the United States,” as used in § 271(a). However, based on a review of the legislative history and a broad construction of § 1498(c)—a construction necessary to limit the scope of Congress’ waiver of sovereign immunity under § 1498(a)—I would hold that § 1498(c) precludes at a minimum those actions premised on infringing activity in which all steps of the infringing process are performed abroad.
Congress enacted § 1498(c) in 1960 as part of its amendments providing an action against the government for copyright infringement. See Pub.L. No. 86-726, 74 Stat. 855 (1960). While Congress considered various drafts, the State Department suggested the language eventually codified at § 1498(c).5 The discussion accompanying the suggestion indicates that the Department specifically intended the language of subsection (c) to remedy concerns that the proposed bill would allow suit for actions abroad, by the United States, which infringed copyrights. See H. Rep. No. 86-624, at 6-7 (1959) (reprinting June 5, 1958 letter from William B. Macomber, Jr., Assistant Secretary of State, to James 0. Eastland, Chairman of the Senate Judiciary Committee). Recognizing that the patent provisions of § 1498(a) had the same ambiguity “with respect to infringements occurring abroad,” the State Department suggested making the foreign country exception applicable to both patent and copyright infringement. Id. at 7. The Commerce Department agreed that there was ambiguity concerning “an act of infringement in a foreign country,” and endorsed the State Department suggestion. See id. at 5-6 (reprinting Feb. 27, 1959 letter from Frederick H. Mueller, Under Secretary of Commerce, to Emanuel Cel-ler, Chairman of the House Judiciary Committee). The foreign country exception at § 1498(c) adopts the State Department suggestion without change. See Pub.L. No. 86-726, 74 Stat. 855 (1960) (codified in relevant part at § 1498(c)). In short, the legislative history confirms that Congress intended § 1498(c) to foreclose actions on § 1498(a) violations taking place in, or arising in, a foreign country.
In construing statutory language, it is prudent to be cognizant of constructions that courts have accorded to identical language, where it occurs in other statutes. Here I recognize that Congress took the language of § 1498(c) from identical language found at 28 U.S.C. § 2680(k), adopted twelve years earlier.6 Normally, where Congress borrows language from one statute in enacting another, the meaning accorded terms in the source statute can be taken as probative of the meaning in the borrowing statute. See Lorillard v. Pons, 434 U.S. 575, 580-81, 98 S.Ct. 866, 55 L.Ed.2d 40 (1978); id. at 581, 98 S.Ct. *1362866 (“[W]here ... Congress adopts a new law incorporating sections of a prior law, Congress normally can be presumed to have had knowledge of the interpretation given to the incorporated law, at least insofar as it affects the new statute[.]”).
Plainly, many determinants of the borrowed statute canon are satisfied here. The statutes serve similar purposes. Both provisions limit the potential government liability occurring abroad by curtailing Congress’s sovereign immunity waiver for tort liability. Moreover, during Senate sub-committee hearings, the testimony made clear that § 1498(c) was intended to provide an exclusion analogous to § 2680(k).7 In this case, however, the meaning of § 2680(k) has limited value for interpreting § 1498(c). This is true because the borrowed statute canon is not absolute, and it does not justify results either inconsistent with the statutory scheme at bar, or leading to functional absurdity in application. Consequently, for the reasons given below, I conclude that the borrowed statute canon does not compel us to interpret § 1498(c) in light of § 2680(k).
In Sosa v. Alvarez-Machain, 542 U.S. 692, 124 S.Ct. 2739, 159 L.Ed.2d 718 (2004), the Supreme Court held that a “claim arising in a foreign country,” under § 2680(k), referred to where tortious injury is suffered. “[T]he FTCA’s foreign country exception bars all claims based on any injury suffered in a foreign country, regardless of where the tortious act or omission occurred.” Id. at 2754. A patent infringement claim, however, is different in kind from the tortious injuries contemplated by § 2680(k).
Though patent infringement claims may be viewed as species of tort claims, they are subject to a territoriality restriction unknown in general tort law. Determining the liabilities of the United States for patent infringement under 28 U.S.C. § 1498(a) requires the application of only United States law, because section 1498 conditions the government’s liability on the existence of patent rights granted by the United States. In contrast, the FTCA requires application of local law to determine the liability of the United States, and under governing choice of law principles, the local law to be applied is typically the law of the place where the injury occurred. 28 U.S.C. § 1346(b)(1) (2000).
Allowing recovery for foreign injury where the causative acts occurred in the United States thus raised the specter that the liability of the United States would be governed by foreign law. No similar problem exists under the patent laws because United States law will apply to the infringing acts occurring in the United States. Under these circumstances, construing the “arising in” language of section 1498(c) to look to the place of the infringing act rather than the place of the injury is per*1363fectly consistent with the Congressional objectives.8
Although Zoltek urges the court to read § 1498(c) as limited to claims arising under foreign patent laws, neither Zoltek’s suggestion nor the Supreme Court’s rationale in Sosa sensibly applies here. First, § 1498(a) plainly provides for actions concerning “invention[s] described in and covered by a patent of the United States[.]” 28 U.S.C. § 1498(a). Zoltek’s analysis would read this provision, in view of § 1498(c), as waiving sovereign immunity to allow infringement actions under a foreign patent if the asserted claim coincided with protections under an issued U.S. patent. I find that reading stylized and unpersuasive. If Congress intended to allow suit against the government for infringing a foreign patent, it would have so stated, and not limited the possible actions by reference to parallel U.S. patents. Moreover, Zoltek’s suggestion is inconsistent with the plain language of § 1498(c) and the structure of § 1498. It would make little sense for Congress to curtail the sovereign immunity waiver at § 1498(c) if the underlying action in § 1498(a) allowed suit “under foreign law.” In short, § 1498 is consistent with a provision to vindicate a U.S. patent right, not a foreign patent right. Second, for similar reasons, unlike § 2680(k), § 1498(a) does not require a choice of law analysis and the Sosa consideration of place of injury is simply inapplicable here. In sum, the § 1498(a) claim arises where the infringing activity is consummated.
This analysis precludes Zoltek’s Nicalon fiber allegations. Where all the steps of a process patent are performed abroad, the claim “arises in a foreign country,” and the action is barred by § 1498(c). The record evidences that the accused Nicalon fiber products were made entirely in Japan. Zoltek’s contention that the government violated § 1498(a) when Lockheed’s Japanese subcontractors “used” Zoltek’s method claims to make the Nicalon products, therefore, fits squarely within our interpretation of § 1498(c). The same holds for the Tyranno fiber allegations, insofar as the fibers were made in Japan. The Court of Federal Claims correctly determined that it lacked jurisdiction to decide those allegations.
2.
The infringement allegations concerning the Tyranno fiber products, however, cannot be tethered to any single location. This presents the difficult question. Although the fibers were made overseas, the record shows that the final products were made in the United States. According to Zoltek’s infringement allegations, some of the claimed method steps were practiced in Japan, and some in the United States. The allegedly infringing conduct, in short, was not confined within national boundaries.
Does such conduct “arise in a foreign country” within the meaning of § 1498(c)? As explained herein I conclude that it *1364does, with the effect of barring Zoltek’s Tyranno fiber product allegations. Put differently, I read “claims arising abroad” expansively, as capturing any claim that does not arise entirely through domestic infringing conduct. In doing so, I reject any contention that the analysis depends on where any individual step of the claimed method was practiced.
Two longstanding interpretative principles support the conclusion that the language at § 1498(c), limiting the sovereign immunity waiver at § 1498(a), includes within its proscription the conduct disputed in this action. First, the courts generally disfavor extraterritorial application of United States law. See EEOC v. Arabian Am. Oil Co., 499 U.S. 244, 248, 111 S.Ct. 1227, 113 L.Ed.2d 274 (1991) (“It is a longstanding principle of American law that legislation of Congress, unless a contrary intent appears, is meant to apply only within the territorial jurisdiction of the United States.”). The principle rests on sound policy, including a concern to avoid conflict with other national laws, regard for international comity, choice of law analysis, and separation of powers. Thus, in Deepsouth, the Supreme Court expressly refused to extend the scope of § 271(a) to capture an extra-territorial manufacture of an infringing device absent express guidance from Congress. “Our patent system makes no claim to extraterritorial effect ... and we correspondingly reject the claims of others to such control over our markets.” 406 U.S. at 531, 92 S.Ct. 1700 (citing Brown v. Duchesne, 60 U.S. (19 How.) 183, 195, 15 L.Ed. 595 (1856) and Boesch v. Graff, 133 U.S. 697, 703, 10 S.Ct. 378, 33 L.Ed. 787 (1890)). In narrowing the reach of actions under § 1498(a), § 1498(c) manifests a Congressional intent to limit infringement actions where conduct by or for the government sounded abroad. Reading § 1498(c) to foreclose the action at bar is consistent with a strict view of the extra-territoriality principle.9
Second, this view comports with the narrow construction that a court must afford sovereign immunity waivers. A court must resolve textual ambiguity in a statutory waiver in favor of immunity. As noted above, the waiver must be clear, and ambiguities resolved in favor of immunity. Since the waiver must be construed narrowly, the § 1498(c) exclusion—serving to narrow the waiver at § 1498(a)—must be construed broadly. Cf. Shoshone Indian, 364 F.3d at 1346. Given an expansive reading, these principles indicate that a claim “arises in a foreign country” under § 1498(c) if, as with the Tyranno product allegations, some steps of a claimed method are practiced abroad.10
Although this court has never decided how § 1498(c) applies to transnational conduct in practicing a patented method, on one occasion our predecessor court was confronted with allegations of infringing use of a transnational device. In Decca Ltd. v. United States, 210 Ct.Cl. 546, 544 F.2d 1070 (1976), as in this case, the government argued that § 1498(c) was a dis-positive bar to jurisdiction. The Court of *1365Claims, however, neither expressly interpreted nor applied the statute.11 Instead, as explained below, it crafted a test for determining an effective “location” of the accused, internationally disposed, device. The Decca court then determined that if such a device could be said to be located “within the United States,” the § 1498(a) claim for infringing use would arise where the device was “located.” For several reasons I conclude that no comparable test should apply to the facts in this case.
To understand why, it bears reviewing the case in detail. Decca involved the Omega long range radio navigation system. Certain components of the system— broadcast stations—were located overseas. The plaintiff, Decca, owned U.S. Patent No. 2,844,816, which claimed radio navigation systems.12 It alleged that the Omega system infringed certain device claims in the ’816 patent and that the government violated § 1498(a) by using the Omega system. The government sought dismissal, arguing that because some broadcast towers were located in Norway and other overseas locations, the action arose in a foreign country and was barred by § 1498(c).
The court disagreed. The Court of Claims defined the “location” of the Omega system to be “in United States territory,” implicitly reasoning that an infringing use “within the United States,” a fortiori, does not give rise to an infringement claim “arising in a foreign country” under § 1498(c). In sum, Decca stands for the narrow, and collateral, proposition that under § 1498(a) the infringing use of a claimed device “arises” where an accused device is “located.”
There are, however, no “use of device” claims at issue in this action, and Decca is therefore inapposite. In Decca, the elements of a patented device were assembled, one by one and in different countries, and the device was therefore made. Likewise, in this case, the steps of a patented process were practiced, one by one and in different countries, and the process was therefore performed. However, these cases are not analogous, because the concept of the “use” of a device is fundamentally different from the use of a process. Whereas utility can be extracted from a device only after it has been “made,” utility is extracted from a process concurrent with its being “practiced.” See NTP, 418 F.3d at 1318 (“Because a process is nothing more than the sequence of actions of which it is comprised, the use of a process necessarily involves doing or performing each of the steps recited. This is unlike use of a system as a whole, in which the components are used collectively, not individually.”). As a consequence of this failure in the analogy, Decca is neither on point nor particularly instructive.
Consequently, I would decline any invitation to extend Decca’s rationale beyond its intended context of device claims to practicing the steps of a claimed method. In particular, the Decca rationale for characterizing the accused Omega radio navi*1366gation system as “located ... within the United States” threatens to swallow the § 1498(c) limitation if applied to the process claim at bar. Decca’s infringement analysis placed “particular emphasis on the ownership of the equipment by the United States, the control of the equipment from the United States and on the actual beneficial use of the system within the United States.” 544 F.2d at 1083. Relying on these observations, both the trial court and the appellate panel concluded that the “location of the infringement is within United States territory.” Id. at 1074. The panel relied, in particular, on the fact that the infringing broadcast equipment “is designed and built by the United States Government,” and that the “location of the whole ... is where the ‘master’ station or stations are, which is in the United States of America, and where all the stations are monitored, presently in Washington, D.C.” Id.
Were a court to extend Decca’s reasoning here, it could conclude that where the government, or a contractor acting at the government’s direction, commissions or directs the conduct later accused of infringing a method patent, the resulting claim for infringing use under § 1498(a) “arises” within the United States, and not abroad, under § 1498(c). Applying the Decca reasoning out of its intended context of device claims, however, would effectively read the plain language of § 1498(c) out of existence. The statute concerns allegations of infringement against the government, and it is difficult to imagine circumstances where such conduct would not be traceable back to the federal government, thus completely avoiding the narrowing effect of § 1498(c) on Congress’s waiver of sovereign immunity. In short, if the accused conduct is always traceable to decision making in the United States, proximately causing otherwise “infringing” conduct abroad, then such claims will never be restricted by the prohibition at § 1498(c).
III.
In sum, the infringing use of a patented method under § 1498(a) consists of practicing all steps in the claimed method. Because the injury from infringement is intangible and not amenable to geographical reference, the cause of action cannot be said to arise where injury is suffered, and the § 2680(k) cases are of limited value in interpreting § 1498(c). I would thus look to where the infringing conduct occurred to ascertain where the claim arises. In the first analysis, no particular step has primacy. But since a court must give the sovereign immunity waiver at § 1498 narrow construction, I must view the exclusion at § 1498(c) broadly and resolve ambiguity in favor of immunity. For that reason, § 1498(c) precludes an action premised on infringing use of a method claim if some steps of the method are practiced abroad. Since Zoltek’s Tyranno product claims falls within this construction, they are barred. Thus, the trial court properly concluded that it lacked jurisdiction over those infringement allegations.
The dissent argues that I must be careful to construe § 1498 in a manner that does not somehow run afoul of the Fifth Amendment. Its concern is fundamentally misguided, however, even if one assumes, for sake of argument, that a patent is a type of property to which Fifth Amendment protections apply. Section 1498 is a waiver of sovereign immunity, the scope of which is completely within the discretion of *1367Congress.13 In fact, there is no Fifth Amendment requirement that § 1498 exist at all; it is the responsibility of Congress, and of Congress alone to decide whether, and to what extent, it will permit the courts to help it fulfill its Constitutional obligations under the Takings Clause.14 A court need not concern itself with whether § 1498, construed narrowly in recognition of Congress’ sole power to waive sovereign immunity, provides a compensation scheme that is compliant with the requirements of the Takings Clause. It is not our place to interpret sovereign immunity waivers with an eye toward their Takings Clause sufficiency vel non, because the Constitution does not require Congress to create remedies in the courts at all. If Congress did not create a statutory scheme for compensation as provided under § 1498, it would follow that without such a provision, compensation would need to be obtained through legislative action.
IV.
For the aforegoing reasons, I believe that there exists an additional, and independent, basis for reaching the decision of the court today that § 1498(a) liability for the use of process claims requires all the steps to have been practiced in the United States. In my view, therefore, our decision today could have been reached without relying on NTP.
. In a § 1498 case, determining the scope of the claims proceeds via the same two-step analysis that occurs in a § 271(a) action, namely claim construction, followed by a determination of whether the claims'—as construed—read on the accused device. See Le-melson v. United States, 752 F.2d 1538, 1548 (Fed.Cir.1985) ("Although this court has noted that a section 1498 action and a title 35 action are only parallel and not identical, the principles of claim construction and reading claims on accused devices and methods are the same for either type of action.”) (citations omitted). Of course, the rights triggered when embodiments fall within the claims may differ under § 1498 and § 271.
. The Court of Federal Claims ultimately concluded that § 1498(c) precluded any claim for a "use" of a patented process unless every step of the claimed process took place "within the United States.” Zoltek, 51 Fed.Cl. at 836.
. See 35 U.S.C. § 271(a)(2000) ("Except as otherwise provided in this title, whoever without authority makes, uses, offers to sell, or sells any patented invention, within the United States or imports into the United States any patented invention during the term of the patent therefor, infringes the patent.").
. Although NTP merges the question of conduct (the use) and location (within the United States) in this explanation of § 271(a), the court plainly adopts the proposition that an infringing use requires practicing every step of a claimed method.
. The State Department comments were directed to H.R. 8419, a predecessor bill to H.R. 4059. H.R. 4059 was enacted on September 8, 1960. See Pub.L. No. 86-726, 74 Stat. 855 (1960) (codified at 28 U.S.C. § 1498).
. The Federal Tort Claims Act ("FTCA”) waives sovereign immunity for certain tort claims against the United States. See 28 U.S.C. § 1346(b). The statute contains a foreign country exception that precludes § 1346(b) liability for "[a]ny claim arising in a foreign country.” 28 U.S.C. § 2680(k). Congress enacted this provision in 1948. See 62 Stat. 984, 985 (1948). Twelve years later, Congress used the same language when enacting § 1498(c). See Pub.L. No. 86-726, 74 Stat. 855 (1960) (codified at § 1498(c)).
. See Infringement of Copyrights: Hearings on H.R. 4059 Before the Senate Subcommittee On Patents, Trademarks, and Copyrights of the Senate Committee on the Judiciary, 86th Cong. 12-13 (1960) (statement of Arthur Fisher, U.S. Registrant of Copyrights) ("[I]t was originally felt that [§ 1498(c)] was unnecessary, that a tort committed abroad would not be subject to suit under the act.... [I]t is quite clear that whatever is the situation with respect to infringements committed abroad, this act leaves the situation in status quo; it does not affect it. [§ 1498(c) ] simply says that this [section] shall not apply to torts committed abroad[.] ... [0]ther agencies have not felt safe in operating under the Government Tort Claims Act, and ... we should simply follow along with a substantially parallel position to the patent law.”).
. Our cases concerning personal jurisdiction in patent cases, which look to "the location ... at which the infringing activity directly impacts on the interests of the patentee,” Beverly Hills Fan Co. v. Sovereign Corp., 21 F.3d 1558, 1571 (Fed.Cir.1994), address where suit may be brought to enforce an existing substantive claim, but those cases do not define the substantive claim itself. The question of whether a patent infringement claim can be maintained at all (controlled by sections 1498(a) and 1498(c) where the alleged in-fringer is the government) is analytically distinct from the personal jurisdiction issue. Any uncertainty as to the scope of the patent law must be determined before addressing personal jurisdiction over the claim.
. The dissent suggests that we have ignored the Fifth Amendment in determining the proper jurisdictional scope provided by § 1498.
. While not a basis for my interpretation of § 1498, I note that any rule that allowed sovereign immunity, and liability, to turn on where a defendant practiced any given step of a claimed method would simply invite strategic manipulation and prove unworkable. Potential infringers would simply move their conduct offshore—if only for the last step of the process—in order to avoid the specter of § 1498(a) liability.
. Indeed, although the United States briefed the issue in conjunction with a motion to dismiss, the trial division of the Court of Claims specifically noted in its pretrial order that ''[jjuriscliction is not disputed and is hereby determined to be present.” Decca, No. 299-70, Pre-Trial Order ¶1 (Ct.Cl. Mar. 20, 1975).
. The trial court found the government liable for an infringing use of claim 11, which claimed a hyperbolic radio navigation system. See '816 patent, col. 12, 1. 63—col. 13, 1. 34 (claim 11); Decca, 544 F.2d at 1083. With minor changes in the factual findings, the three judge panel adopted the trial court’s analysis. Decca, 544 F.2d at 1072.
. See Lynch v. United States, 292 U.S. 571, 582, 54 S.Ct. 840, 78 L.Ed. 1434 (1934) ("The sovereign’s immunity from suit exists whatever the character of the proceeding or the source of the right sought to be enforced.... [including] those arising from some violation of rights conferred upon the citizen by the Constitution ....” (citing Schillingerv. United States, 155 U.S. at 166, 168, 15 S.Ct. 85, 39 L.Ed. 108 ("The United States cannot be sued in their courts without their consent, and in granting such consent Congress has an absolute discretion to specify the cases and contingencies in which the liability of the Government is submitted to the courts for judicial determination. Beyond the letter of such consent the courts may not go, no matter how beneficial they may deem, or in fact might be, their possession of a larger jurisdiction over the liabilities of the Government.”))).
. Indeed, before the passage of the Tucker Act, ch. 359, § 1, 24 Stat. 505 (1887) (codified as amended at 28 U.S.C. § 1491), Congress had sole responsibility for paying takings claims. No judicial relief was available. See, e.g., Langford v. United States, 101 U.S. 341, 343, 25 L.Ed. 1010 (1879) (holding that notwithstanding the obligation of the United States to pay just compensation, there existed no remedy in the courts because "Congress has made no provision by any general law for ascertaining and paying this just compensation”). In Lynch, v. United States, the Supreme Court reaffirmed the discretion of Congress to reserve to itself the payment of just compensation for Fifth Amendment takings. 292 U.S. 571, 54 S.Ct. 840, 78 L.Ed. 1434. In that case, the Court started by recognizing that "[r]ights against the United States arising out of a contract with it are protected by the Fifth Amendment.” Id. at 579, 54 S.Ct. 840. It held, however, that Congress may elect to retain the sole power to remediate alleged "takings” of these contractual rights. See id. at 582, 54 S.Ct. 840 ("When the United States creates rights in individuals against itself, it is under no obligation to provide a remedy through the courts. It may limit the individual to administrative remedies. And withdrawal of all remedy, administrative as well as legal, would not necessarily imply repudiation. So long as the contractual obligation is recognized, Congress may direct its fulfillment without the interposition of either a court or an administrative tribunal." (emphasis added) (citing United States v. Babcock, 250 U.S. 328, 331, 39 S.Ct. 464, 63 L.Ed. 1011 (1919) and Tutun v. United States, 270 U.S. 568, 576, 46 S.Ct. 425, 70 L.Ed. 738 (1926))).