Passamaquoddy Tribe v. State of Maine

UNITED STATES COURT OF APPEALS UNITED STATES COURT OF APPEALS FOR THE FIRST CIRCUIT FOR THE FIRST CIRCUIT No. 95-1922 PASSAMAQUODDY TRIBE, Plaintiff, Appellant, v. STATE OF MAINE, ET AL., Defendants, Appellees. APPEAL FROM THE UNITED STATES DISTRICT COURT FOR THE DISTRICT OF MAINE [Hon. Morton A. Brody, U.S. District Judge] Before Selya, Boudin and Lynch, Circuit Judges. Thomas N. Tureen, with whom Gregory W. Sample, Tureen & Sample, Richard B. Collins, David Overlock Stewart, and Ropes & Gray were on brief, for appellant. Francis A. Brown on brief for City of Calais, Maine, amicus curiae. Thomas D. Warren, Assistant Attorney General, with whom Andrew Ketterer, Attorney General, and Wayne Moss, Assistant Attorney General, were on brief, for appellees. February 9, 1996 SELYA, Circuit Judge. The Passamaquoddy Tribe (the SELYA, Circuit Judge. Tribe) sued to compel Maine and the governor of Maine (collectively, Maine or the State) to recognize its asserted right to avoid the prohibitions of Maine's criminal code, see 17- A Me. Rev. Stat. Ann. 953-954, and conduct high-stakes casino gambling behind the shield of the Indian Gaming Regulatory Act, 25 U.S.C. 2701-2721, 18 U.S.C. 1166-1168 (the Gaming Act). The federal district court decided that the Gaming Act does not extend to Maine, and denied relief. See Passamaquoddy Tribe v. Maine, 897 F. Supp. 632 (D. Me. 1995). We affirm. I. THE STATUTORY FRAMEWORK I. THE STATUTORY FRAMEWORK In order to put this appeal into perspective, it is necessary to juxtapose the Gaming Act and the Maine Indian Claims Settlement Act of 1980, 25 U.S.C. 1721-1735 (the Settlement Act). In the early 1970s, the Tribe began earnestly to pursue claims to nearly two-thirds of Maine's land mass. See Joint Tribal Council of the Passamaquoddy Tribe v. Morton, 388 F. Supp. 649, 651-53, 667-69 (D. Me.) (reviewing dispute's history), aff'd, 528 F.2d 370 (1st Cir. 1975). After years of strife, the Tribe and the State negotiated a settlement of the land claims under federal auspices. The arrangement was designed to transform the legal status of the Maine tribes (the Passamaquoddy Tribe and the Penobscot Nation), and to create a unique relationship between state and tribal authority. See Penobscot Nation v. Stilphen, 461 A.2d 478, 488-89 (Me.), appeal dismissed, 2 464 U.S. 923 (1983). The Passamaquoddies and the Penobscots ratified the provisional pact and Maine's legislature followed suit. See P.L. 1979, c. 732, codified at 30 Me. Rev. Stat. Ann. 6201-6214. In 1980, Congress cemented the terms of the accord by passing the Settlement Act. The federal statute incorporated the parties' agreement and established the ground rules that henceforth would govern matters of common political concern to the State and the two tribes. Among other things, the Settlement Act rid the State of all Indian land claims and submitted the Passamaquoddies, the Penobscots, and their tribal lands to the State's jurisdiction. See 25 U.S.C. 1721(b)(4), 1723(b) & (c), 1725(a). In addition, section 16(b) of the Settlement Act gave the State a measure of security against future federal incursions upon these hard-won gains. It stated: The provisions of any federal law enacted after October 10, 1980 [the effective date of the Settlement Act], for the benefit of Indians, Indian nations, or tribes or bands of Indians, which would affect or preempt the application of the laws of the State of Maine, . . . shall not apply within the State of Maine, unless such provision of such subsequently enacted Federal law is specifically made applicable within the State of Maine. 25 U.S.C. 1735(b) (emphasis supplied). The Tribe received fair consideration for its agreement: the Settlement Act confirmed its title to designated reservation lands, memorialized federal recognition of its tribal status, and opened the floodgate for the influx of millions of dollars in federal subsidies. See 25 3 U.S.C. 1733. Approximately eight years later, Congress enacted the Gaming Act. This statute establishes a three-tiered regulatory paradigm in respect to gambling activities on Indian lands. We described these three layers in Rhode Island v. Narragansett Indian Tribe, 19 F.3d 685, 689-90 (1st Cir.), cert. denied, 115 S. Ct. 298 (1994), and it would be pleonastic to rehearse that description here. We focus instead on the third tier: Class III gaming (a category that encompasses casino gambling). The Gaming Act provides that, unless a state imposes an outright ban on all Class III gaming (and Maine does not), it must, upon the request of a federally recognized and self- governing Indian tribe, negotiate a compact stipulating the terms and conditions under which the tribe can introduce Class III gaming on Indian lands. See 25 U.S.C. 2710(d). The statute contains a series of fail-safe mechanisms designed to ensure that states do not stall the negotiations or conduct them in bad faith. See, e.g., id. 2710(d)(7). The Settlement Act and the Gaming Act are vastly different in scope. From a geographic standpoint, the former is narrower in the sense that it applies only in Maine whereas the latter has national implications. From a political standpoint, however, the Settlement Act is broader in that it purposes to cover virtually the entire field of relationships between the State and the Indian tribes based there whereas the Gaming Act concentrates exclusively on a particular kind of activity, i.e., 4 gambling. II. THE GENESIS OF THE APPEAL II. THE GENESIS OF THE APPEAL Mindful of the meteoric success of other Indian- sponsored casinos, the Tribe decided in the early 1990s to climb aboard the bandwagon. It chose Calais, a Maine municipality located near the Canadian border, as the preferred site for its nascent enterprise. Because the Gaming Act requires Class III gaming to be conducted on "Indian lands," 25 U.S.C. 2710(d)(3)(A), the Tribe sought to add a designated parcel of real estate to its inventory of tribal lands. See 30 Me. Rev. Stat. Ann. 6205 (authorizing incremental land acquisitions). When formally apprised of the Tribe's plans, the State concluded that the Gaming Act did not apply within Maine's boundaries and scotched the proposed casino. As a lagniappe, the state legislature passed a bill that allowed tribal land in Calais to be used for such a purpose (1) if the Tribe secured the city's blessing and the Governor of Maine thereafter agreed to negotiate a tribal-state compact under 25 U.S.C. 2710(d), or (2) if a court of competent jurisdiction declared that the Gaming Act extended to Maine. See Me. Laws 1993, ch. 713, 1, codified at 30 Me. Rev. Stat. Ann. 6205(1)(c). After some procedural maneuvering, not material here, the Tribe sued to compel the commencement of negotiations for a compact. The defendants moved for judgment on the pleadings, Fed. R. Civ. P. 12(c), asserting that the Gaming Act did not hold 5 sway within Maine. The Tribe opposed the motion. It contended among other things that the Gaming Act reached Maine, as elsewhere, because Congress had impliedly repealed the Settlement Act vis-a-vis gaming activities conducted by Indian tribes, and, in all events, had made the Gaming Act specifically applicable within Maine. Unimpressed by the Tribe's armada of arguments, the district court ruled that the Gaming Act lacked force in Maine and entered judgment in the defendants' favor. See Passamaquoddy Tribe, 897 F. Supp. at 635. This appeal followed. III. ANALYSIS III. ANALYSIS Our discussion of the issues proceeds in four parts. A A This case turns on a question of statutory interpretation. By its terms, the Gaming Act, if taken in isolation, applies to any federally recognized Indian tribe that possesses powers of self-governance. See 25 U.S.C. 2703(5). Consequently, if we were to start and stop with the Gaming Act, the Tribe which is federally recognized and self-governing would be home free. But this case cannot be confined within such narrow margins. The chief objective of statutory interpretation is to give effect to the legislative will. See Negonsott v. Samuels, 113 S. Ct. 1119, 1122-23 (1993); Narragansett Indian Tribe, 19 F.3d at 691. To achieve this objective a court must take into account the tacit assumptions that underlie a legislative enactment, including not only general policies but 6 also preexisting statutory provisions. See Ohio ex rel. Popovici v. Agler, 280 U.S. 379, 383 (1929); Greenwood Trust Co. v. Massachusetts, 971 F.2d 818, 827 (1st Cir. 1992), cert. denied, 113 S. Ct. 974 (1993). Put simply, courts must recognize that Congress does not legislate in a vacuum. See Thinking Machines Corp. v. Mellon Fin. Servs. Corp. # 1 (In re Thinking Machines), 67 F.3d 1021, 1025 (1st Cir. 1995). Taking this haploscopic view brings us immediately to section 16(b) of the Settlement Act, 25 U.S.C. 1735(b), quoted supra p.3. At first glance, the conditions precedent to the applicability of section 16(b) are plainly satisfied. The Tribe does not dispute nor could it that the Gaming Act is a "federal law enacted after October 10, 1980, for the benefit of Indians, Indian nations, or tribes or bands of Indians, which would affect or preempt the application of the laws of the State of Maine."1 25 U.S.C. 1735(b). In such circumstances, section 16(b) provides that Maine will be exempt from such a statute unless Congress has "specifically made" the statute "applicable within the State of Maine." In other words, section 16(b) is a savings clause that serves two related purposes. It acts as a warning signal to later Congresses to stop, look, and listen before weakening the foundation on which the settlement between Maine and the Tribe rests. At the same time, it signals courts that, if a later Congress enacts a law for the benefit of 1Among other things, the Gaming Act, if it applied, would preempt various provisions of Maine's criminal law, including 17- A Me. Rev. Stat. Ann. 953-954. 7 Indians and intends the law to have effect within Maine, that intent will be made manifest. In view of these dual purposes, we cannot decide the question of whether the Gaming Act extends to Maine withoutfactoring section 16(b) intothe decisional calculus. This realization gets the grease from the goose. The text of the Gaming Act contains not so much as a hint that Congress intended to make that Act specifically applicable within Maine. Where, as here, Congress enacts a statute of general applicability (e.g., the Gaming Act) with full knowledge that a preexisting statute (e.g., the Settlement Act) contains a savings clause warning pointedly that a specific reference or a similarly clear expression of legislative intent will be required to alter the status quo, the only reasonable conclusion that can be drawn from the later Congress's decision to omit any such expression from the text of the new statute is that Congress did not desire to bring about such an alteration. See Narragansett Indian Tribe, 19 F.3d at 704 n.21 (observing that when an "enacting Congress is demonstrably aware of the earlier law at the time of the later law's enactment, there is no basis for indulging" any other presumption). The Tribe's principal rejoinder is on constitutional grounds. It posits that giving effect to section 16(b) in this fashion is tantamount to binding a successor Congress to a predecessor's will, and therefore careens beyond the constitutional pale. See, e.g., Glidden Co. v. Zdanok, 370 U.S. 530, 534 (1962); Reichelderfer v. Quinn, 287 U.S. 315, 318 8 (1932); Fletcher v. Peck, 10 U.S. (6 Cranch) 87, 135 (1810). We believe that this rejoinder distorts the reality of events. Section 16(b) does not prohibit a subsequent Congress from writing a new statute reflecting new policies and applying it to the Indian tribes in Maine. Congress could make such a statute fully effective in Maine through the use of explicit language, by otherwise offering a patent indication of its intent to accomplish that result, or, indeed, by first repealing section 16(b). Thus, section 16(b) is purely an interpretive aid; it serves both to limn the manner in which subsequently enacted statutes should be written to accomplish a particular goal and to color the way in which such statutes thereafter should be read. In fine, section 16(b) binds subsequent Congresses only to the extent that they choose to be bound. The sockdolager is that the Court regularly has upheld and given effect to such provisions, see, e.g., Warden, Lewisburg Penit. v. Marrero, 417 U.S. 653, 659-60 n.10 (1974) (earlier statute barred repeal of certain penalties "unless the repealing Act shall so expressly provide"); Shaughnessy v. Pedreiro, 349 U.S. 48, 52 (1955) (earlier statute directed that "[n]o subsequent legislation shall . . . supersede or modify the provisions of [the earlier statute] except to the extent such legislation shall do so expressly"); Posadas v. National City Bank, 296 U.S. 497, 501 (1936) (earlier statute directed that subsequent laws "shall not apply to the Philippine Islands, except when they specifically so provide"); Great Northern Ry. 9 Co. v. United States, 208 U.S. 452, 456 (1908) (similar to Marrero); United States v. Reisinger, 128 U.S. 398, 401-02 (1888) (similar to Marrero), and we see nothing that distinguishes this case from the mine-run. This means, of course, that we must read the Settlement Act and the Gaming Act in pari passu. Doing so, and giving effect to their plain meaning, we are led inexorably to the conclusion that the latter lacks force within Maine's boundaries. B B The Tribe generates several other responses to our tentative conclusion that Congress did not intend to make the Gaming Act operative in Maine. Its most ferocious attack suggests that section 16(b) need not be considered at all because the Gaming Act impliedly repealed it insofar as gambling on tribal lands is concerned. The attack is easily repulsed. We are unequivocally committed to "the bedrock principle that implied repeals of federal statutes are disfavored." Narragansett Indian Tribe, 19 F.3d at 703; accord Rodriguez v. United States, 480 U.S. 522, 524 (1987); TVA v. Hill, 437 U.S. 153, 189 (1978); United States v. Borden Co., 308 U.S. 188, 198 (1939). The general rule is that "when two statutes are capable of coexistence, it is the duty of the courts, absent a clearly expressed congressional intent to the contrary, to regard each as effective." Morton v. Mancari, 417 U.S. 535, 551 (1974). The only other satisfactory basis for a repeal by implication (apart from a clear expression of 10 Congress's intent to repeal) is a finding that the earlier and later statutes are irreconcilable. See Hill, 437 U.S. at 190; Morton, 417 U.S. at 550; Narragansett Indian Tribe, 19 F.3d at 703-04. "[I]f the two [acts] are repugnant in any of their provisions, the latter act, without any repealing clause, operates to the extent of the repugnancy as a repeal of the first." United States v. Tynen, 78 U.S. (11 Wall.) 88, 92 (1870). Of course, statutes can be irreconcilable even short of outright repugnancy. Thus, a repeal may be implied if a later statute covers the entire subject matter "and embraces new provisions, plainly showing that it was intended as a substitute for the first act." Id.; see also Posadas, 296 U.S. at 503-04; Narragansett Indian Tribe, 19 F.3d at 703-04. But an irreconcilable conflict does not exist merely because the application of a later statute would "produce differing results when applied . . ., for that no more than states the problem." Radzanower v. Touche Ross & Co., 426 U.S. 148, 155 (1976). These precepts fit without special tailoring in the Indian law context. See, e.g., Narragansett Indian Tribe, 19 F.3d at 704; Blackfeet Indian Tribe v. Montana Power Co., 838 F.2d 1055, 1058 (9th Cir.), cert. denied, 488 U.S. 828 (1988). In this case, they defeat the Tribe's attack. The Gaming Act contains no evidence of an intention to repeal section 16(b) of the Settlement Act, let alone a patent expression of any such design. Indeed, when the 100th Congress passed the Gaming Act it 11 was fully cognizant of the Settlement Act and apparently contemplated that the new statute would not in any way displace the old: [I]t is the intention of the Committee that nothing in . . . [the Gaming Act] will supersede any specific restriction or specific grant of Federal authority or jurisdiction to a State which may be encompassed in another Federal statute, including the . . . [Maine] Indian Claim[s] Settlement Act. S. Rep. No. 446, 100th Congress, 2d Sess. 12 (1988), reprinted in 1988 U.S.C.C.A.N. 3071, 3082.2 The absence of any suggestive guideposts in the Gaming Act, coupled with the easy integration of the two laws, effectively dispatches the argument for implied repeal. Our opinion in Narragansett Indian Tribe is not to the contrary. There, we concluded that Congress, in passing the Gaming Act, had impliedly repealed the Rhode Island Indian Claims Settlement Act of 1978, 25 U.S.C. 1701-1716, to the extent that it touched upon gambling activities. See Narragansett 2We found this passage of no help in the context of the Rhode Island Indian Claims Settlement Act of 1978. See Narragansett Indian Tribe, 19 F.3d at 700. The version of the bill to which the report applied originally contained a provision that explicitly exempted Rhode Island from the reach of the Gaming Act, yet, prior to enactment, Congress removed the exonerative provision. In that circumstance, we concluded that the report "shed[] no light on Congress's intent regarding the law it actually enacted." Id. By contrast, the draft bill appended to the report did not contain any similar language regarding Maine (presumably because the legislators knew that the Settlement Act included a savings clause making such language unnecessary). Thus, unlike in the case of Rhode Island, no telltale chain of events taints the report's reference in respect to Maine. 12 Indian Tribe, 19 F.3d at 704-05. But the Rhode Island Act contained no provision comparable to section 16(b); therefore, the literal terms of the two statutes created incoherence by subjecting Indian gaming to two mutually exclusive regulatory environments. Because we could find no feasible way to give full effect to both acts, we concluded that an implied repeal had transpired. See id. Here, in contradistinction to the situation that obtained in Rhode Island, section 16(b) satisfactorily harmonizes the Settlement Act and the Gaming Act, and prevents any incoherence. The Settlement Act governs the State's relationship with the Tribe and will continue to do so without dilution unless and until Congress, by later enactment, makes a new law touching upon the same subject matter in one or more particulars specifically applicable within Maine. As the Gaming Act does not meet this benchmark, the Settlement Act remains inviolate and precludes the operation of the Gaming Act in Maine. See Ysleta del Sur Pueblo v. Texas, 36 F.3d 1325, 1335 (5th Cir. 1994) (holding that the Gaming Act did not impliedly repeal a federal statute granting Texas jurisdiction over Indian gaming because Congress never indicated in the Gaming Act that it intended to rescind the previous grant of jurisdiction), cert. denied, 115 S. Ct. 1358 (1995). To sum up, we do not find it surprising that the lack of any express indicium of a contrary congressional intent in the text of the Gaming Act means different things in different 13 settings. Without a savings clause like section 16(b), this omission may indicate an intent to apply the Act across the board especially if, as in Narragansett Indian Tribe, Congress weighed, and decided to discard, a specific exemption. But when a savings clause is in play, as in this case, the omission can only mean that Congress desired the terms of the earlier statute to prevail. In the final analysis, the differing outcomes in the two New England states bear witness to the truism that, "[i]n the game of statutory interpretation, statutory language is the ultimate trump card." Narragansett Indian Tribe, 19 F.3d at 699. C C The Tribe has a fallback position. It maintains that, even if we give full force and effect to section 16(b), the Gaming Act controls because it is "specifically made applicable" within Maine. In its most primitive form, this thesis embodies a contention that because the Tribe satisfies the Gaming Act's general definitional requirement federal recognition and governmental power a court can infer Congress's intent to bestow the benefices of the Gaming Act upon the Tribe. The problem with this contention is that it entirely ignores the Settlement Act. Once that flaw is revealed, it becomes readily apparent that the Tribe's contention is no more than a back-door effort to reintroduce the notion of implied repeal. Consequently, we reject it. 14 In a related vein, the Tribe postulates that the very comprehensiveness of the Gaming Act is itself enough to meet the demands of section 16(b). This asseveration depends heavily upon the correctness of the proposition that the rule of Marcello v. Bonds, 349 U.S. 302 (1955), permits minimal particularity of expression to satisfy savings clauses like section 16(b). We do not believe that the proposition withstands scrutiny. In Marcello, a provision of the Administrative Procedure Act (APA) stipulated that statutes which purport either to supersede or modify the APA's judicial review modalities must do so "expressly." See id. at 305 (quoting APA 12, now codified at 5 U.S.C. 559). A later Congress enacted the Immigration and Nationality Act of 1952 (I&N Act). Although the I&N Act did not override the APA's judicial review modalities in so many words, the Supreme Court concluded that the neoteric statute's deportation procedure superseded the APA's judicial review modalities because (1) the presence in the I&N Act of an extensive review scheme, similar in material respects to the APA's review mechanisms, would otherwise be rendered meaningless, and (2) the I&N Act contained an explicit provision that the procedure which it prescribed "shall be the sole and exclusive procedure for determining the deportability of an alien." See Marcello, 349 U.S. at 308-09. These factors, together with some instructive legislative history, formed the basis for the Court's determination that the subsequent Congress had "expressly" 15 superseded the APA's judicial review modalities in respect to deportation.3 Id. at 310. The Tribe's reliance on Marcello is mislaid. To be sure, the Gaming Act, like the I&N Act, is a statute of general applicability that arguably constructs a comprehensive regulatory regime for a defined subject.4 But this single similarity does not provide a particularly persuasive parallel for present 3The Court wrote that it could not ignore the background of the 1952 immigration legislation, its laborious adaptation of the Administrative Procedure Act to the deportation process, the specific points at which deviations from the Administrative Procedure Act were made, the recognition in the legislative history of this adaptive technique and of the particular deviations, and the direction in the statute that the methods therein prescribed shall be the sole and exclusive procedure for deportation proceedings. Marcello, 349 U.S. at 310. The Court then concluded: Unless we are to require the Congress to employ magical passwords in order to effectuate an exemption from the Administrative Procedure Act, we must hold that the present statute expressly supersedes the hearing provisions of that Act. Id. 4The State argues that the Gaming Act is not comprehensive in the conventional sense. This argument is not totally without merit; the Gaming Act has no application to tribes that do not seek and attain formal federal recognition, see 25 U.S.C. 2703(5), tribes that do not exercise jurisdiction over their territories, see id. 2710(b)(1) & (d)(3)(A), tribal lands located in states that proscribe Class II and III gaming activities altogether, see id. 2710(b)(1) & (d)(1), or tribal lands on which federal law pretermits gambling, see id. 2710(b)(1). We need not probe the point too deeply. For present purposes, we simply assume, favorably to the Tribe, that the Gaming Act, like the I&N Act, constitutes a comprehensive regulatory regime. 16 purposes. Here, the Tribe points to nothing of consequence beyond the comprehensive nature of the Gaming Act. Unlike the deportation procedure delineated in the I&N Act, none of the provisions of the Gaming Act will be rendered meaningless if the Act excludes Maine. Moreover, unlike in the I&N Act, Congress has not declared the Gaming Act to be "exclusive" of other potentially applicable legislation. And, finally, unlike in the legislative history of the I&N Act, there are no signposts writ large in the debate over the Gaming Act. These differences serve both to distinguish the instant case from Marcello and to put the holding of that case into perspective. See Great Northern, 208 U.S. at 466 (explaining that the comprehensiveness of subsequent legislation, without more, will not satisfy a savings clause in an earlier statute). The point is not that Congress was derelict in employing one particular collocation of words as opposed to another, but, rather, that it chose not to include in the Gaming Act any indication that it meant to make the statute specifically applicable within Maine.5 Though their arguments are unavailing when weighed on an evenly calibrated scale, the Tribe seeks to tip the balance by 5We find puzzling the Tribe's reliance on a line of cases, see, e.g. Sims v. CIA, 471 U.S. 159, 167 (1985), decided under an exemption from the disclosure provisions of the Freedom of Information Act, 5 U.S.C. 552(b)(3) (providing that agencies need not divulge matters that are "specifically exempted" by statute), to support its ipse dixit that Congress need only enact a comprehensive statute to mute the call of section 16(b). That exemption merely incorporates by reference the secrecy provisions of other statutes, and, unlike section 16(b), plays no discernible role in construing the application of a subsequently enacted statute. 17 altering the calibration. To this end, it invites us to depart from the usual canons of construction and chart the statutory interface between the Gaming Act and the Settlement Act by resort to a special interpretive preference that the law sometimes accords to Indian tribes. See, e.g., Amoco Prod'n Co. v. Village of Gambell, 480 U.S. 531, 555 (1987); South Carolina v. Catawba Indian Band, Inc., 476 U.S. 498, 506 (1986) (collecting cases); Rosebud Sioux Tribe v. Kneip, 430 U.S. 584, 586-87 (1977). We decline the invitation. The rule of construction to which the Tribe alludes reflects a strong federal interest in safeguarding Indian autonomy. See, e.g., Rosebud Sioux, 430 U.S. at 586-87. But the rule is apposite only when Congress has blown an uncertain trumpet. If ambiguity does not loom, the occasion for preferential interpretation never arises. See Catawba Indian Band, 476 U.S. at 506; Rosebud Sioux, 430 U.S. at 587-88; Narragansett Indian Tribe, 19 F.3d at 691. When, as now, Congress has unambiguously expressed its intent through its choice of statutory language, courts must read the relevant laws according to their unvarnished meaning, without any judicial embroidery. So it is here: since there is no statutory ambiguity, the principle of preferential construction is not triggered. D D The Tribe's last argument has a different spin. Under the Gaming Act, Class II gaming conducted on tribal lands must be 18 sanctioned by the National Indian Gaming Commission. See 25 U.S.C. 2710(b). While this litigation was pending, the Tribe adopted an ordinance authorizing the conduct of bingo and other Class II gaming activities on its reservation lands and submitted this proposal to the Commission. The Commission asserted jurisdiction and granted the request. The approval took the form of a letter dated July 19, 1995, in which the Commission's chairman opined that the Gaming Act applied in Maine. The Tribe asked the district court to take judicial notice of, and defer to, that determination. See generally Chevron U.S.A. Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837, 842-43 (1984) (discussing deference due to agency interpretations); Strickland v. Commissioner, Me. Dep't of Human Servs., 48 F.3d 12, 16 (1st Cir.) (similar), cert. denied, 116 S. Ct. 145 (1995). The district court demurred. The Tribe assigns error. We discern none. It is transpicuously clear that, under Chevron, no deference is due if Congress has spoken directly to the question. See Strickland, 48 F.3d at 16. Here, we read section 16(b) of the Settlement Act as a clear and unambiguous expression of congressional intent. Furthermore, in light of section 16(b), the Gaming Act's failure to mention Maine makes that statute, too, compelling evidence of Congress's intent that it should not apply in Maine.6 6The Tribe construes the Gaming Act's silence as a latent ambiguity. We do not agree. Given the tenor of the preexisting statute, the sound of silence here is pregnant with meaning. 19 In this instance, moreover, there is another valid reason for declining to defer to the Commission. Deference is appropriate under Chevron only when an agency interprets a statute that it administers. See CFTC v. Schor, 478 U.S. 833, 845 (1986). Here, the question of the Gaming Act's applicability cannot be addressed in a vacuum, and the Commission, whatever else might be its prerogatives, does not administer the Settlement Act. That role belongs to the Secretary of the Interior, see, e.g., 25 U.S.C. 1725, 1727(a), and has not been delegated by the Secretary to the Commission. Though the Commission may have expertise in the conduct of gaming activities on tribal lands, see, e.g., Shakopee Mdewakanton Sioux Community v. Hope, 16 F.3d 261, 264 (8th Cir. 1994), we cannot take it upon ourselves to assume, without any evidence, that Congress intended to entrust the Commission with reconciling the Gaming Act and other statutes in the legislative firmament. If more were needed and we do not believe that it is we note that deference is inappropriate when an agency's conclusion rests predominantly upon its reading of judicial decisions. See, e.g., Director, OWCP v. General Dynamics Corp., 980 F.2d 74, 78-79 (1st Cir. 1992). In this instance, the Commission's jurisdictional analysis depends almost exclusively on decrypting and applying Marcello and Narragansett Indian Tribe. As courts, not agencies, have special expertise in Taken in context, that silence logically denotes Congress's intent not to make the Gaming Act specifically applicable within Maine. 20 interpreting case law, we are loath to defer to a determination that amounts to little more than the Commission's understanding of judicial precedents. IV. CONCLUSION IV. CONCLUSION To recapitulate, the Tribe and the State negotiated the accord that is now memorialized in the Settlement Act as a covenant to govern their future relations. Maine received valuable consideration for the accord, including the protection afforded by section 16(b). The Tribe also received valuable consideration, including land, money, and recognition. Having reaped the benefits, the Tribe cannot expect the corollary burdens imposed under the Settlement Act to disappear merely because they have become inconvenient. We need go no further. We hold that Congress did not make the Gaming Act specifically applicable within Maine, and that, therefore, the Tribe is not entitled to an order compelling the State to negotiate a compact for Class III gaming. Affirmed. Affirmed 21