No. 98-377: Dickinson v. Zurko

No. 98-377

In the Supreme Court of the United States OCTOBER TERM, 1998 Q. TODD DICKINSON, ACTING COMMISSIONER OF PATENTS AND TRADEMARKS, PETITIONER v. MARY E. ZURKO, ET AL. ON WRIT OF CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE FEDERAL CIRCUIT REPLY BRIEF FOR THE PETITIONER SETH P. WAXMAN Solicitor General Counsel of Record ALBIN F. DROST Department of Justice Acting Solicitor Washington, D.C. 20530-0001 United States Patent and (202) 514-2217 Trademark Office Arlington, VA 22215 In the Supreme Court of the United States OCTOBER TERM, 1998 No. 98-377 Q. TODD DICKINSON, ACTING COMMISSIONER OF PATENTS AND TRADEMARKS, PETITIONER v. MARY E. ZURKO, ET AL. ON WRIT OF CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE FEDERAL CIRCUIT REPLY BRIEF FOR THE PETITIONER 1. Respondents base their argument on the contention (Br. 9-13) that, before the enactment of the APA in 1946, courts reviewed factual determinations made by the Patent Office under a well-established standard, equivalent to the "clearly erroneous" standard that is presently used to review district court factual findings under Rule 52(a) of the Federal Rules of Civil Procedure. The purported existence of such a pre-APA standard is thus critical to respondents' position, but there are at least two significant flaws in their historical analysis. First, respondents' argument depends on whether a "clear error" standard of judicial review was clearly established and consistently applied before the enactment of the APA. Yet even the court of appeals conceded that "[i]t would be disingenuous to suggest that the courts employed a uniform standard of review prior to 1947." Pet. App. 11a; see also id. at 15a-16a. That acknowledged ambiguity undermines respondents' argument that a standard of review equivalent to the modern "clear error" rule was so clearly "recognized by law" as to be preserved, by Section 12 of the APA (now 5 U.S.C. 559), as an exception to the uniform standards otherwise prescribed by that Act for the judicial review of administrative action. Second, respondents rely too blithely on parenthetical snippets from opinions written 50 or more years ago. Resp. Br. 9-13. Although many of those decisions use variations on the words "clear" and "error," or similar phrases, the contextual significance of the language respondents quote is not self-evident. Respondents make no effort to demonstrate that the decisions they cite actually applied a level of scrutiny that would be best characterized as review for "clear error," rather than review under a "substantial evidence" standard, as those terms of art are presently understood. Of the 36 pre-APA cases cited by respondents (Br. 9-13, 1a- 2a), only four actually reversed the PTO. See Resp. Br. 12 (citing Townsend v. Smith, 36 F.2d 292 (C.C.P.A. 1929); Reusch v. Fischer, 49 F.2d 818 (C.C.P.A. 1931); Kreidel v. Parker, 97 F.2d 171 (C.C.P.A. 1938); and In re Herchenrider, 117 F.2d 261 (C.C.P.A. 1941)). In three of those cases (Townsend, Reusch and Kreidel), the court explicitly subjected the PTO's decision to de novo review-a standard that respondents contend was applied in only "a few isolated cases" (Br. 12), and which they do not advocate. In the fourth case (Herchenrider), the court, after a careful review of the prior art in relation to the claims at issue, resolved its own remaining doubts as to patentability in favor of the applicant using a presumption of patentability that, according to respondents, was applied "occasionally," but was "subsequently rejected in favor of the clearly erroneous standard." Ibid. Thus, respondents cite no case in which a PTO determination was reversed as "clearly erroneous," but might have been upheld as supported by "substantial evidence."1 Moreover, the pre-APA cases that respondents cite frequently stress the PTO's technical expertise as a primary reason for employing a deferential standard of review. In In re Hornsey, 48 F.2d 911 (C.C.P.A. 1931), for example, the PTO concluded-much as it did in the present case-that a patent applicant's combination of various elements, all of which were previously known in the art, was not sufficiently original to be patentable. In sustaining that determination, the Court of Customs and Patent Appeals explained: The process necessarily involves a consideration of highly technical matters. The Board and the Examiner have given these matters careful consideration, and, notwithstanding appellant's exhaustive treatment of the subject, we are not convinced that the Board's conclusions are wrong. In matters appealed here from the Patent Office, which involve the findings of the experts of the Office on highly technical questions, such findings are given great weight and are only rejected when it is clear that they are erroneous. Since it is admitted that all of the elements of the process are old, we are not prepared to say, contrary to the decision of the Board, that invention resulted from the combination of these old elements. Id. at 912 (citations and paragraph break omitted). The key point of this passage is not the court's use of "clear" in proximity to "erroneous," but rather the principle that it is appropriate for the court to defer to the PTO's expertise in determining whether a patent application does or does not describe a patentable advance over the prior state of a technical art.2 That principle has little or nothing to do with the "clearly erroneous" standard applied by courts of appeals in reviewing district court factual findings. District courts do not possess differential substantive expertise that warrants deference by a reviewing court. Differential expertise is, on the other hand, an important aspect of the justification for the sort of deferential review that courts have accorded to the findings of specialized administrative agencies under the "substantial evidence" standard. See 5 U.S.C. 706(2)(E); see also, e.g., Universal Camera Corp. v. NLRB, 340 U.S. 474, 488 (1951) (agencies are "presumably equipped or informed by experience to deal with a specialized field of knowledge," and, under the "substantial evidence" test, their "findings within that field carry the authority of an expertness which courts do not possess and therefore must respect"); Abbott v. Coe, 109 F.2d 449, 451- 452 (D.C. Cir. 1939) (noting that principles of deferential review "have special force when the administrative tribunal of the Patent Office has decided a technical question within its field, for 'it is just such questions that the administrative tribunal is pre-eminently qualified to solve,'" and dealing separately with the principle that a trial court's findings will not be disturbed unless "clearly wrong"); Stern, Review of Findings of Administrators, Judges and Juries: A Comparative Analysis, 58 Harv. L. Rev. 70, 80-83 (1944) (distinguishing "clearly erroneous" and "substantial evidence" standards in part on this basis). Thus, the cases respondents cite for their pre-APA "rule" often bear at least as close a functional resemblance to the administrative-review cases that developed the "substantial evidence" standard, which Congress incorporated into the APA (see id. at 74-77), as to the equity cases that developed the "clearly erroneous" standard now embodied in Rule 52(a) (see id. at 79-80, 86-87). A final example decisively underscores the overconfidence of respondents' assertions concerning the nature of pre-APA review. As noted above, respondents cite In re Herchenrider as an example of a rule that was "occasionally" applied by the Court of Customs and Patent Appeals to resolve doubtful cases in favor of patentability. Resp. Br. 12. They then cite General Motors Corp. v. Coe, 120 F.2d 736, cert. denied, 314 U.S. 688 (1941), a case decided by the Court of Appeals for the District of Columbia, for the proposition that that rule "was subsequently rejected in favor of the clearly erroneous standard." Resp. Br. 12. General Motors did decline to apply the Herchenrider rule (although with a potential procedural reservation), and it reiterated the "settled law" that the court would "sustain the findings of the Patent Office and the District Court unless [it thought] them clearly wrong." 120 F.2d at 737. In its next sentence, however, the court reformulated that point, in terms that suggest caution in reading its previous statement as a simple declaration of reliance on the "clearly erroneous" standard now embodied in Rule 52(a): "In other words[,] doubt is to be resolved, in suits to obtain a patent as in suits of other sorts, in favor of the correctness of administrative and judicial action." Ibid. That need for caution is emphatically confirmed by the fact that, in declining to resolve doubts in favor of patentability, the author of the General Motors opinion quoted directly (120 F.2d at 737 & n.5) from his own opinion for the court some six months earlier in Minnesota Mining & Manufacturing Co. v. Coe, 118 F.2d 593 (D.C. Cir. 1940), cert. denied, 314 U.S. 624 (1941). In Minnesota Mining, the court sustained the rejection of certain patent claims, on the ground that the PTO's determination that the claims were "not patentable over the prior art" (id. at 593) was supported by "[s]ubstantial evidence." Id. at 595-595 (emphasis added). There is no ambiguity in that language, because the court explained its standard by quoting Consolidated Edison Co. v. NLRB, 305 U.S. 197, 229 (1938)-perhaps the leading case from this Court defining the "substantial evidence" standard that was later incorporated into the APA. 118 F.2d at 594 & n.12; see, e.g., U.S. Dep't of Justice, Attorney General's Manual on the Administrative Procedure Act 109 (1947) (citing Consolidated Edison); Resp. Br. 13, 15 (same). Moreover, a few years later-and just three months before President Truman signed the APA-the same court reiterated its test that "a reasonable finding that claims lack invention should not be set aside." Besser v. Ooms, 154 F.2d 17, 18 (D.C. Cir. 1946) (emphasis added). In so doing, it explicitly noted that, although it had also stated that it would reverse factual findings if they were "clearly wrong," that "review formula [was] accurate only with respect to judicial, not administrative, findings." Id. at 18 n.3 (citing cases). See also Abbott v. Coe, 109 F.2d at 451 ("The question for [the court] is not whether in our opinion there was invention, but whether the finding that there was none is consistent with the evidence."); compare Resp. Br. 14-15 (review of PTO factfinding "has always been at the level of clear error review and never [under] the more deferential substantial evidence standard").3 2. For similar reasons, respondents can derive little comfort from their observation (Br. 14) that the enactment of the APA caused no apparent change in the judicial review of PTO determinations. To the contrary, as a source cited by respondents themselves makes clear, there was never any doubt that the provisions of the new Act-including those dealing with judicial review-applied to the PTO. See Zitver, The Administrative Procedure Act, 28 J. Pat. Off. Soc'y 676 (1946); see also Ooms, The United States Patent Office and the Administrative Procedure Act, 38 Trademark Rep. 149 (1948). If the standards of review prescribed by the APA had been perceived as significantly different from those previously in use, then surely courts and practitioners would have noted that fact, and either adapted their practices or explained, at that time, why they would continue to adhere to different rules, despite their facial inconsistency with the new Act. Yet respondents cite no such discussion, and we are aware of none. Under these circumstances, a contemporaneous statement like that cited by respondents, that "Other sections of the act apply to the Office only in the sense that they embody existing practices, for example, the section on judicial review" (Resp. Br. 37, quoting Zitver, 28 J. Pat. Off. Soc'y at 676) bears only one reasonable interpretation: That the standards of review prescribed by the new Act, while applicable in challenges to PTO actions, were not viewed as requiring any substantial change in past practice. The lack of apparent reaction from judges and practitioners after the passage of the APA thus confirms our point that respondents and the Federal Circuit have signally failed to demonstrate that pre-APA cases required the application of a "clear error" standard of review that would have been understood, at the time, as materially more demanding than the standards ultimately prescribed by the APA. 3. Respondents point out (Br. 15-16) that a legal rule established by judicial decision (rather than by statute) could be "otherwise recognized by law" within the meaning of 5 U.S.C. 559. We have not argued otherwise. See Gov't Br. 24-25. As respondents grudgingly concede (see Br. 16 n.14), however, the word "recognized" places some limit on what non-statutory "requirements" Section 559 may be read to preserve. Our submission (Br. 24-25) is merely what common sense and ordinary usage would suggest: That the full statutory phrase at issue, "imposed by statute or otherwise recognized by law," implies a roughly equal dignity between potential sources of "additional requirements," so that a common law (or administrative) rule should not be treated as "recognized" for purposes of Section 559 unless it was so clearly established, at the time the APA was adopted, that it could fairly be compared with a rule "imposed by statute."4 In light of our discussion of the pre-APA cases cited by respondents (pp. 2-8, supra), and the court of appeals' concession (see Pet. App. 11a, 15a) that the pre-APA cases established no "clear" or "uniform" standard of review, respondents have failed to establish that any standard of review less deferential than those prescribed by the APA was "recognized by law" at the time the APA was enacted. 4. If we nonetheless assume the existence of some such preexisting, more stringent standard of review, then the issue is whether that standard was preserved by the "additional requirements" language in the first sentence of what is now 5 U.S.C. 559. As we explain in our opening brief (at 21-26), respondents' argument on that point rests on an unpersuasive reading of the statutory text. Respondents dispute our reading of the statutory term "requirements" (Gov't Br. 22-24) primarily by arguing that the "judicial review provisions" of the APA are not categorically "exempted from the reach" of Section 559. Resp. Br. 20-22; see also id. at 32-33. That is true, but unresponsive. Our point is not that Section 559 somehow exempts the Act's review provisions, or that the word "requirements" could not possibly be read to include standards of review. We argue only that, in the context of Section 559, that word -which appears not only in the first sentence, at issue here, but also in the second and third sentences of the present Section, and three times in the last sentence of the corresponding Section of the original Act (see Gov't Br. 2-3, 24)-is better read to refer to the sort of affirmative obligations that the Act imposes directly on covered agencies, or on those who invoke the agency's procedures to seek or oppose agency action. Cf., e.g., King v. St. Vincent's Hosp., 502 U.S. 215, 221 (1991) ("[T]he meaning of statutory language, plain or not, depends on context."); Estate of Cowart v. Nicklos Drilling Co., 505 U.S. 469, 478-479 (1992) (it is a "basic canon of statutory construction that identical terms within an Act bear the same meaning"). Respondents argue further (Br. 19-20, 23-24) that if a standard of review is a "requirement[]," then the term "additional" should be read broadly to cover any non-APA standard that is "more searching, demanding, burdensome, onerous, rigorous, stringent or strict" than any prescribed by the Act itself. There is, however, a fundamental distinction between a non-APA rule that merely supplements (or "add[s]" to) any "requirement[]" imposed by the Act-such as, for instance, the additional statutory "hearing" requirement that was at issue in United States v. Florida East Coast Railway, 410 U.S. 224 (1973)-and a non-APA rule that addresses in a different way an issue that is otherwise comprehensively treated by the Act itself (and thus displaces, rather than adds to, the Act's "requirements"). A construction of the word "additional" that respects that distinction represents the more logical contextual interpretation of Section 559.5 5. Accordingly, this Court's decision in Florida East Coast Railway is entirely consistent with our interpretation of Section 559. As noted above (and in our opening brief at 26), Florida East Coast Railway, unlike this case, involved a pre-APA statutory directive that clearly imposed a procedural "requirement[]" of the sort contemplated by Section 559 (i.e., that the agency act only "after [a] hearing"). See 410 U.S. at 225-226 & n.1, 234-235. That requirement, as the Court construed it (id. at 234-238), also plainly supplemented, rather than either invoked or contradicted, the "hearing" provisions of the APA itself.6 6. Respondents correctly observe (Br. 36) that the APA emerged from many years of active debate, a good deal of which "focused on the intensity of judicial review." As this Court has put the point, the Act "represent[ed] a long period of study and strife; it settle[d] long-continued and hard-fought contentions, and enact[ed] a formula upon which opposing social and political forces [had] come to rest." Wong Yang Sung v. McGrath, 339 U.S. 33, 40 (1950). Contrary to respondents' argument (Br. 34-40), however, that history of contention and compromise supports the uniform application, in judicial review of all agency decisions, of the standards of review that Congress ultimately chose to embody in the APA itself. For example, respondents cite a recent study of the APA's legislative history for the proposition that Carl McFarland of the American Bar Association (ABA) ultimately "acknowledg[ed] that the final compromise provisions on judicial review 'merely confirmed existing case law.'" Resp. Br. 36-37 (quoting Shepherd, Fierce Compromise: The Administrative Procedure Act Emerges from New Deal Politics, 90 Nw. U. L. Rev. 1557, 1660 (1996)). But that citation is seriously incomplete. According to Professor Shepherd's detailed discussion, although there was some confusion on the point, it was understood at the time the APA was debated and enacted that the prevailing verbal standard of judicial review was the "substantial evidence" standard articulated by this Court in cases such as Consolidated Edison. See 90 Nw. U. L. Rev. at 1600, 1602, 1621 & n.314, 1660, 1664; see also Universal Camera, 340 U.S. at 477-484. From the beginning of the debate over administrative reform, one goal of those concerned by the growth of administrative power was to subject agency action-and particularly agency factfinding-to more stringent judicial review. See 90 Nw. U. L. Rev. at 1573, 1591-1593, 1597-1598, 1613-1614, 1624, 1636-1637, 1644-1645, 1657, 1664-1665, 1680.7 Professor Shepherd explains, however, that by the time of the "final compromise[s]" (Resp. Br. 37) in drafting the APA, those who favored changing the law to require stricter review had realized that they lacked the necessary votes. See 90 Nw. U. L. Rev. at 1655-1657, 1659-1660, 1663-1666. Especially in light of that background, the passage cited by respondents actually supports uniform application of the relatively deferential "substantial evidence" standard: Likewise, the ABA retreated from its demand for broad judicial review. The ABA no longer sought to permit a reviewing court to reweigh evidence under the 'preponderance of the evidence' standard - 'it would cause about as much difficulty as help.' Instead, the bar now contented itself with the Senate draft's 'substantial evidence' rule, which McFarland conceded merely confirmed existing case law. Id. at 1660 (emphasis added; footnotes omitted).8 The centrality, from the beginning, of issues of judicial review in the years-long debate over administrative reform that culminated with enactment of the APA also belies respondents' attempt (Br. 39-40) to minimize our point (Gov't Br. 30-31) that those deeply involved in that debate specifically considered, but ultimately rejected, inclusion of the "clearly erroneous" standard among those to be legislatively prescribed. Given the attention that such issues received during the extended efforts to craft a legislative framework for the regularization and control of administrative action, that considered decision, at a critical point in the history of the legislation (during the final crafting and passage of the Walter-Logan bill), is highly probative in illuminating the final legislative compromise embodied in the APA. See Stern, 58 Harv. L. Rev. at 87-88 & n.73 (discussing the legislative debate over inclusion of the "clearly erroneous" provision, and noting that the controversy "sheds considerable light on the meaning of both the 'clearly erroneous' and 'substantial evidence' tests"); Shepherd, 90 Nw. U. L. Rev. at 1621 (noting the "heated debate * * * about the provision for the clearly erroneous standard").9 In view of these aspects of the APA's history, it is not enough for respondents to insist (Br. 34-38) that the judicial review provisions of the APA were intended merely to "restate" the law of judicial review. As we have indicated (Gov't Br. 29), that is true as a general matter, and if the phrase is properly understood. But it does not advance respondents' position. First, as the passage we have quoted from Professor Shepherd's article makes clear, the law that Congress thought it was "restating" is-unsurprisingly-the law that is set out (or "restated") in the Act. Second, as we have seen (see pp. 1-10, supra), respondents have produced no evidence to suggest that the standards prescribed by the APA were perceived at the time as markedly different from some other standard that had been applied consistently in pre-Act patent cases. Finally, even if there had been some marginal inconsistency (such as, for example, an occasional administrative decision that might have struck the court as "clearly wrong," but would nonetheless have been upheld if the applicable test clearly required only "substantial evidence"), respondents wholly fail to explain why the APA's "restatement" would not have been intended, and should not be construed, to eliminate such minor deviations from the norm. See Gov't Br. 29-30. Certainly such a construction is the one most consistent with the Act's overarching purpose of securing "stricter and more uniform practice." Universal Camera, 340 U.S. at 489 (emphasis added); see also ibid. (APA's adoption of "substantial evidence" test was "not a reflection of approval of all existing practices"); Wong Yang Sung, 339 U.S. at 41 (an important purpose of the APA was "to introduce greater uniformity of procedure and standardization of administrative practice among the diverse agencies whose customs had departed widely from each other"). 7. Finally, even if the inquiry were a pertinent one, respondents fail to offer any sound reason to depart from the APA's prescribed standards of judicial review in the context at issue in this case. Respondents suggest three policy reasons for refusing to apply the APA's standards in reviewing PTO decisions denying patent applications. First, they second the Federal Circuit's claim that applying a more stringent standard of review will produce better administrative decisions. Resp. Br. 41 (citing Pet. App. 25a). If, however, the court of appeals concludes that the administrative decision in a particular case is inadequate to permit proper review, then the appropriate remedy is presumably to remand the case to the agency, specifying the material defects and requiring that they be corrected so that review may proceed-not to change the applicable standard of review. Cf. Motor Vehicle Mfrs. Ass'n v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 40-44, 48-51, 57 (1983) (same standard of review applies to adoption and later rescission of agency rule, but agency must consider relevant factors and adequately explain its decisions); SEC v. Chenery Corp., 332 U.S. 194, 196-197 (1947) (agency must adequately explain its decision; if court cannot sustain action on grounds given, proper remedy is to remand to the agency for further consideration). Moreover, it is unclear how applying a more stringent standard of review to patent denials-ordinarily resulting in the subsequent issuance of patents without any alteration of the record on which review was performed-would "encourage administrative records that more fully describe the metes and bounds of the patent grant" (Pet. App. 25a), or otherwise produce any result other than the issuance, on supposedly inadequate records, of patents that the expert agency created by Congress to consider such matters has concluded should not be issued. Respondents next argue (Br. 42-43) that application of the proper APA standard in review of administrative patent denials would result in a "two-standard scheme," because factual findings made by district courts in review proceedings brought under 35 U.S.C. 145 would presumably be reviewed for "clear error." See Fed. R. Civ. P. 52(a). Whether any such difference would actually "skew the review process, impose undue burdens on applicants [or] inevitably lead to irreconcilable results" (Resp. Br. 42) is open to considerable doubt.10 As we explain in our opening brief (at 34-36), however, any anomaly in that arrangement simply mirrors the usual differences in appellate review of judicial and administrative factfinding, and results from Congress's historical choice to give disappointed patent applicants the option either to seek review on the administrative record in the court of appeals, or to initiate an independent judicial review proceeding in the district court, with the opportunity to expand the record, and then a further opportunity for appellate review. Third, respondents contend that the PTO's determinations should be subject to non-APA review because its procedures for considering and acting on patent applications differ in some respects from those that other agencies use for other purposes. Resp. Br. 43-44; see generally Gov't Br. 6-9 (describing agency procedures). As the PTO has consistently argued (and as respondents and the Federal Circuit now agree), it is an "agency" subject to the requirements of the APA; and the procedures that respondents cite could therefore be challenged if they failed to conform to any applicable norm prescribed by Congress. As respondents implicitly concede, however, the agency's practices are consistent with the statutory framework prescribed by the patent laws, and they violate no provision of the APA. The present argument therefore reduces to the assertion that because Congress has not imposed on the PTO procedural requirements that respondents would deem desirable, this Court should decline to apply other provisions of the APA in accordance with their terms. That is not a persuasive proposition. Tellingly, although respondents adduce these policy rationales to support the result reached by the court of appeals, they do not defend (or even mention) the central justification advanced by that court for its result: That it must be free to subject the PTO's decisions to "heightened * * * scrutiny" so that it may review them "on [the court's] own reasoning," rather than on that adopted by the agency. Pet. App. 3a, 25a-27a; see Resp. Br. 44-45; compare Patent, Trademark & Copyright Section of the Bar Ass'n of D.C. Amicus Br. 4-5, 15, 24 (explicitly defending this position). As we have explained (Gov't Br. 38-41), the court's acknowledged rationale for applying a non-APA standard of review makes clear that its decision in this case exceeds the proper limits of judicial review of administrative action. Accordingly, this case ultimately involves the proper balance between the powers and responsibilities of the Federal Circuit, as a reviewing court, and the expert administrative agency to which Congress has committed primary responsibility for the administration of the Nation's patent laws. That balance was struck by Congress, after long debate, when it enacted the APA; and it may be maintained simply-but only-by enforcing that Act in accordance with its terms. * * * * * For the foregoing reasons, and those set forth in our opening brief, the judgment of the court of appeals should be reversed. Respectfully submitted. SETH P. WAXMAN Solicitor General ALBIN F. DROST Acting Solicitor United States Patent and Trademark Office FEBRUARY 1999 1 The New York Intellectual Property Law Association identifies 90 pre-APA cases that purportedly applied a "clear error" standard. NYIPLA Amicus Br. 4-5 & 1a-6a. Our review of those cases again discloses no case in which reversal of a PTO finding appears to turn on the use of a "clear error" rather than a "substantial evidence" standard. See, e.g., In re Breer, 55 F.2d 485, 486 (C.C.P.A. 1932) (applying the presumption of patentability that respondents do not defend); In re Engelhardt, 40 F.2d 760, 761-762 (C.C.P.A. 1930) (indulging what amounts to a strong presumption of patentability, under "extraordinary" procedural circumstances, while allowing for further dispute on that issue in future infringement litigation). 2 See also, e.g., In re Ruzicka, 150 F.2d 550, 553 (C.C.P.A. 1945) ("The subject matter involves a highly technical chemical question, and it would necessarily have to be clear that the board erred in this respect before we would be warranted in reversing its holding."); In re Ubbelohde, 128 F.2d 453, 456 (C.C.P.A. 1942) ("The question of what would be obvious to one skilled in the art before us involves the consideration of extremely technical matters; under these circumstances we would not be warranted in reversing the Patent Office tribunals unless we believe that those tribunals are manifestly wrong. This we cannot do upon the record before us."); In re Batcher, 59 F.2d 461, 463 (C.C.P.A. 1932) ("[T]he presumption [of expertise] applies, and * * * when an appeal is taken to this court, the judges of which are not supposed to be, and do not profess to be, experts in the realm of mechanics, the burden rests upon the party appealing to make it clear that the findings of fact by such tribunals are manifestly wrong."); Rowe v. Holtz, 55 F.2d 468, 470-471 (C.C.P.A. 1932) ("The question involves highly technical matters of electrical engineering, and we cannot say that the tribunals below were clearly in error in their findings."); In re Wietzel, 39 F.2d 669, 671 (C.C.P.A. 1930) ("In cases involving intricate and highly technical questions, * * * concurring decisions of the Patent Office tribunals will not be disturbed, unless it appears that they are manifestly wrong."); In re Ford, 38 F.2d 525, 526 (C.C.P.A. 1930) ("[W]hen patentable novelty has been denied by all the expert tribunals of the Patent Office, it is incumbent upon one appealing therefrom to make out a clear case of error to obtain a reversal."); cf. Bonine v. Bliss, 259 F. 989, 989-990 (D.C. Cir. 1919) ("The question of whether or not an application is allowable, and one upon which the issuance of a patent can be predicated, is primarily for the experts of the Patent Office, and will not be inquired into in this sort of a proceeding except for manifest error."). 3 If there were any ground for doubt, it would concern whether the "substantial evidence" standard applied in Minnesota Mining was more lenient than the standard ultimately embodied in the APA, because the latter specifically requires that the existence of substantial evidence be evaluated on the "whole record." 5 U.S.C. 706; see Universal Camera, 340 U.S. at 477- 490. 4 The provision that any subsequent statute may be held to supersede or modify the APA only if it does so "expressly" (5 U.S.C. 559) reenforces the point, because it underscores that the APA was intended to put in place a basic, uniform framework that was not to be varied lightly. Of course, this discussion is relevant to resolution of the present case only if a non-APA standard of judicial review of agency action is properly viewed as an "additional requirement[]" within the meaning of Section 559. See pp. 10-16, infra. 5 Compare AT&T v. Central Office Tel., Inc., 118 S. Ct. 1956, 1965 (1998) (a statutory saving clause "cannot in reason be construed as continuing * * * a common law right, the continued existence of which would be absolutely inconsistent with the provisions of the act. In other words, the act cannot be held to destroy itself.") (quoting Texas & Pac. Ry. v. Abilene Cotton Oil Co., 204 U.S. 426, 446 (1907)). 6 SEC v. Morgan, Lewis & Bockius, 209 F.2d 44 (3d Cir. 1953), on which respondents also seek to rely (Br. 27-28), similarly supports our construction of the Act. Morgan, Lewis demonstrates a proper application of Section 559, to rebuff an argument that the APA was intended to occupy the whole field of administrative procedure, preempting any preexisting rule that was not specifically reiterated in the Act itself, even if the rule was not inconsistent with any of the Act's own requirements. See Gov't Br. 25 & n.6. Moreover, as our discussion of these cases makes clear, there is no substance to respondents' contention (Br. 27-28) that our construction of Section 559 would "deny this saving clause any force." See also Abbott Lab. v. Gardner, 387 U.S. 136, 139-144 (1967) (discussing a pre-APA statutory mechanism providing an additional mechanism for judicial review of certain agency actions). 7 Indeed, the only disagreement concerning the existing standard of review apparently came from those who argued, perhaps rhetorically, that under existing law the courts were sustaining agency action if there was even a "scintilla" of evidence to support it. See Shepherd, 90 Nw. U. L. Rev. at 1602, 1636. 8 Proponents of stricter review did succeed in inserting into the Act the explicit requirement that courts "review the whole record," 5 U.S.C. 706, which may have represented an incremental tightening of the common law "substantial evidence" standard. See Universal Camera, 340 U.S. at 477-478, 481-484, 490. To the extent that it did, however, it marks the outer boundary of what proponents of stricter review were able to achieve. 9 Mr. Stern's article, on which respondents themselves rely (Br. 13), also makes clear that respondents err in asserting (Br. 40) that the "clearly erroneous" test was deleted from the bill "before it was considered on the floor of Congress." See Stern, 58 Harv. L. Rev. at 88 n.73 (explaining that the provision for use of that test was debated on the floor of the House (which refused to delete it), deleted by the Senate Judiciary Committee (with an explanation offered during floor debate in the Senate), and explicitly lamented on the floor of the House, although that chamber ultimately concurred in the Senate's action). 10 In particular, there is little realistic possibility that the Federal Circuit would ever be "compelled to hold the same patent both invalid and not invalid over the same prior art simply because of the differing standards of review." Resp. Br. 42 (quoting In re Lueders, 111 F.3d 1569, 1577 (Fed. Cir. 1997)). The potential for rulings on the same patent under different standards would arise only if a claim included in an issued patent were reexamined and cancelled by the PTO, see 35 U.S.C. 301-307 (1994 & Supp. II 1996), but upheld by a district court in infringement litigation-and if the patent applicant opted for review of the former determination on the administrative record in the court of appeals rather than in a district court proceeding, see 35 U.S.C. 306 (conferring option). Moreover, the Federal Circuit itself has previously denied petitioner the right to stay reexamination proceedings pending the decision of a district court in infringement litigation, precisely because any "awkwardness presumed to result if the PTO and court reached different conclusions is more apparent than real. The two forums take different approaches in determining invalidity and on the same evidence could quite correctly come to different conclusions. Furthermore, * * * the district court and the PTO can consider different evidence. Accordingly, different results between the two forums may be entirely reasonable. And, * * * of course, the two forums have different standards of proof for determining invalidity." Ethicon, Inc. v. Quigg, 849 F.2d 1422, 1428-1429 (Fed. Cir. 1988).